Hi, I bought used Giesen W1a couple of weeks ago. I have an issue with drying phase on this machine.
I can't heat the core of the bean, so the bean stays green inside, and during the roasting it turns darker in color than outside of the bean.
This causes unpleasant greenish bitterness.
I am charging at 200c, 1 minute soak at 25% gas and than 90% gas until 150c. Low air until 150c, adding more air after drying and before FC. Curve attached - blue is BT, red is exhaust temp and yellow is environmental temp.
I am not as experienced, can you point me in right direction? I would slow down drying phase a little, using less heat after soak and I would probably try charge hotter.

not seeing your curve, but at what time are you getting yellow and what time/temp are you hitting 1C?

Kenji wrote:Hi, I bought used Giesen W1a couple of weeks ago. I have an issue with drying phase on this machine.
I can't heat the core of the bean, so the bean stays green inside, and during the roasting it turns darker in color than outside of the bean.
This causes unpleasant greenish bitterness.
I am charging at 200c, 1 minute soak at 25% gas and than 90% gas until 150c. Low air until 150c, adding more air after drying and before FC. Curve attached - blue is BT, red is exhaust temp and yellow is environmental temp.
I am not as experienced, can you point me in right direction? I would slow down drying phase a little, using less heat after soak and I would probably try charge hotter.

Try charging at a lower temp, try 185C. It may seem counterintuitive but I'd think charging lower will give the core time to catch up and develop.

I have a 1# USRC and have been charging at 330F to 355F (165C to 180C). I originally started at 380F to 400F and experience the same issues you're having.

chris_n wrote:not seeing your curve, but at what time are you getting yellow and what time/temp are you hitting 1C?

Sorry my fault.

Batch size?

Probe size?

The more I roast the more tricky I find it commenting on graphs from other roasters

That said, my best guess would be, with the same time to yellow, aim for first crack start 9:45 - 10.00, using less air throughout the roast = keep your initial fan setting until just before or just as first crack gets rolling. Lower air will mean you will have to use less gas coming into first crack.

Delta ROR looks pretty flat coming into first crack and then it drops off into first crack pretty hard. I am guessing some roasters would just look at that and call it baked if following RAO. Lately where I am at with my own roasting on a much smaller machine, I would find it hard to disagree.

second thoughts....I agree with Moxi as well, try charging 180 - 185, this might mean a slight reduction on batch size to keep same dry end?

Third thoughts.... based on making 1 change to this roast only = at about 6:45 I would start lowing gas more, that's where the upward trend starts.

All the best for you future roasts.

Thanks for your reply
I am roasting 1kg of coffee in 1,5kg drum.
Probes are 3mm J-type hooked via cropster databridge to Cropster.

Drying looks ok to me. I'm thinking the development is too short and/or not hot enough. So going from 1c to final temp in 2 minutes is not the same as getting there in 3 minutes. Also some beans just need to be roasted hotter than others. I would first try stretching out the roast after 1c and if that doesn't work, then start going hotter. Maybe someone else knows better.

Congrats to your newly acquired roaster and welcome to H-B. Several years ago when I was searching for a commercial sample roaster, the Giesen W1a was on my short list. I was really impressed by its design and functionality. This is the first time I've seen a Giesen user on H-B. If you don't mind, would you share your experience (in a different thread) when you get the hang of it?

As far as I understand, this Giesen W1a can be switched to PID controlled mode, based on the feedback readout from its ET probe? I'm curious about the combined gas and airflow auto-adjustment. When you setup your target ET temperature, can you fix the airflow parameter and the machine would automatically adjust the gas to reach the ET target, or the machine would automatically change both airflow and gas parameters at the same time? How did you test your airflow settings? A cigarette lighter test at the tryer's port? Since you got it used, have you checked the exhaust hose and the cyclone and made sure no dust/gum blocking the vent?

Your roasting profile seems generally good to me, and your probe is accurate and responsive. Have you ever roasted on other machines? How were those machines compared? (For example, if you have used your old air popper to roast the same green bean in an 11-minute profile and get good result, we can then safely rule out the quality of green bean being the culprit.) What is the origin of the bean and do you record % weight loss? Can you stretch the development time (the time after the first crack start) longer (to say 3 minutes) and drop at higher temperature? You have one roast underdeveloped. Try to get another overly done, and your preferred target will be somewhere in between.

Giesens are well insulated and trap heat like an mf-er

But that being said you need to shed a little more heat between yellow and crack. the ROR crash at first is indicative to me that it's probably actually stalling or baking and not sufficiently developing. if you come into first with a Rao-recommended declining ROR you may have more success.

Or you can try his new trick of cutting the heat in half about min before first crack for 20s then back up to where you were before and don't touch until 0:45s into crack.

Either way, you probably don't need to charge as hot, give it more flame at the start, pull back as you go so you enter crack without crashing.

archipelago wrote:Either way, you probably don't need to charge as hot, give it more flame at the start, pull back as you go so you enter crack without crashing.

Agreed. Since you cranked the gas at 1 min to 90%, it looks like too much air happened at that same instance.

Edit: Is that RoR in C/30 sec? If so, RoR may be fine.