Sample roasting > translating to bigger batch

Hello peeps.
I get my Bullet next week, really excited!
I’m trying to do a lot of homework beforehand, and one thing that interests me is the idea of buying 1kg of a bean, then experimenting with 100g or 150g sample batches to see how the flavours change with different profiles.
All very nice, but from what I gather, there is NO way of translating a small batch to a large batch. The profiles will be very different. I don’t entirely understand why… I mean, I understand that more beans will require more heat to keep the same ROR, but IF you managed to keep the curves the same, why would the two roasts not crack at the same time, not look or taste the same?? This baffles me. Maybe it is impossible to control the Bullet in such a way that the curves would be identical with different batch sizes?

Anyway, I would really like to know how I would be able to gain useful information from experimenting with sample batches, if I cannot get the same “best tasting profile” when I roast 750g or so…

I’d love to hear what you pros say, I’ve only been roasting beans for less than a year and have not had a machine that can control much at all… Thanks!


Hi Mark -

Congratulations on your purchase!

I think it is possible to adjust the profile to match the IBTS temperature curves between two roasts of different batch sizes, within a reasonable range of roasting times. I’ve seen that the IBTS crack temperature is pretty independent of batch size, so if you match the curves you should reach the start of first crack at the same time.

One effect of batch size is that the beans don’t interact with each other and with the BT temperature probe in the same way. Beans can be heated by the drum and by contact with other beans. Airflow around individual beans also changes with batch size. I’ve seen that larger batches produce more chaff, so higher fan speeds may be needed to remove the chaff from the drum. Those are a few factors that I think make the flavors in the cup change with batch size.

Let us know what you think of these ideas as you make your own experiments with your new Bullet!


  • Brad

HI Brad, thanks for all the info.
Well, that’s really good news… so “all” I have to do is figure how to steer 3 or 4 separate wheels at the same time while racing down a hill towards a cliff… LOL… but at least I could get there with practice.
To change from a great small load profile to a larger load, I’m guessing that:

1/ I’d need to start with a higher charge temperature to overcome the greater amount of lower temp beans?

2/ Start with a higher Power setting?

3/ Start with slower fan initially to allow heat to build up more easily but increase after yellowing to remove chaff?

Those are (scantly educated) guesses…?


Two problems that we might run into with higher preheats and power are scorching and tipping. Check out this great video by Rob HOOS (scorching @ 15:15, tipping @ 32:00)


Thanks Brad, I’ll give that a watch.