15 Sep 2023

Rob Hoos’ 2023 IKAWA Roast Profile

Rob Hoos has been sharing his profiles with us for about 7 years now. He was first featured in our blog in 2016 and his roast profiles on our Roast Profile Library have been favourites ever since.

Recently we caught up, and we both had the same idea: revisit the roast profile he had shared in 2017 and see if he would update it. It’s no surprise that he’s learned a lot about roasting on IKAWA sample roasters since then, so not only did he update the profile he recommends, he wrote a whole blog about it.

This is reposted with permission from Rob Hoos.

Parting Thoughts on IKAWA PROV3

by Rob Hoos
September 2023

It’s been ages since I first sent my profile to the IKAWA team and did a write-up for their blog. The original write-up was from 2016 when I used an IKAWA PRO V2. A profile for a Kenyan coffee was posted as an update in 2017 (still on the PRO V2). Since then, I have upgraded to the PRO V3 and taught an introductory and advanced IKAWA class (available here on VIMEO). During those courses, I did a deep dive into how I currently understand and approach roasting on the IKAWA… but that was in December 2020. As you may imagine, things have changed, and I am constantly working on refining my roasting on the IKAWA.

As it turns out, this week is the end of an era for me as I say goodbye to my IKAWA PRO V3 and prepare to say hello to an IKAWA PRO100x. Before I upgrade to this fantastic new model, I want to summarize my use of the PRO50 V3 and leave an updated roast profile to use for sample roasting. I don’t feel like my original exhaust profile is something I would use much anymore, and I’d rather leave a different legacy.


ProV3 IKAWA in Rob Hoos workshop

Let’s start with the apparent differences between the IKAWA and what most coffee roasters are accustomed to (drum roasters). My previous statement hints at a distinction: coffee roasting machines are typically classified into two categories. The delineation of these categories comes from the question of what is causing the mixing/agitation of the coffee during the roasting process to ensure even heat application.

The first category is mechanical agitation (which most people are familiar with from the perspective of production roasting). This category of coffee roasters (including drum roasters) uses mechanical means to stir, mix, and loft the coffee. This is frequently a roasting drum with paddles inside the drum chamber that cause the coffee to be mixed and lofted. Other roasters also use mechanical agitation, such as the Stronghold coffee roaster, despite a non-horizontally oriented drum. Another example of a roaster in this category that is often mischaracterized are Loring Coffee Roasters.

The second category of coffee roasters uses airflow (which causes the bean mass to act as a fluid and flow) to move and mix the coffee. These coffee roasters are predominately called fluidized bed coffee roasters (sometimes shortened to fluid beds). This category of roasters is intensely recognizable to people from a home-roasting background. Indeed, many who started on a popcorn popper, heat-gun/bowl combo, iRoast2, Fresh Roast, or Sonofresco are very familiar with this category of roasters. Production roasters may also be familiar with this category by using SIVITZ roasters, Coffee Crafters (or the like). The IKAWA is a fluid bed roaster. It is critical to understand that this category of roasters relies on REMARKABLY more airflow volumes than a mechanical-agitation roaster. When comparing the IKAWA with drum roasters from the perspective of airflow volume per gram of coffee (CFM/g), we can see that there is 6x the airflow in an IKAWA than a Loring s70 at 100%. This is significant and changes the way the coffee roasts. Additionally (at least in the case of the IKAWA), since all of the airflow is flowing up through the coffee, it also changes our expectations of the temperature readings.

illustration of airflow through the IKAWA sample roaster

Since, unlike a drum roasting system where some air bypasses any coffee, a fluidized bed roaster passes all of the airflow through the coffee mass, the bean mass absorbs more energy from the hot air flow. This means that while the inlet air temperature will remain the hottest temperature reading, as in a drum roaster, the exhaust air temperature and the bean temperature measurements will switch positions. Generally, in a drum roaster, the inlet air temperature is the hottest, followed by the exhaust air temperature (because so much of it bypasses coming into contact with the coffee), followed by the bean temperature. In the IKAWA, the bean temperature is the second hottest temperature reading, and the exhaust temperature tends to be the lowest. This runs contrary to my assumptions in the 2016 blog post but is the truth of my findings when I jury-rigged my IKAWA with a bean probe while building my courses.

graph of inlet, exhaust, and bean temperatures from the IKAWA sample roaster

The second important feature of how air flows through the roaster is that the considerable volume of air (relative to the grams of coffee) will force its will onto the coffee. This means that whatever natural path the coffee bean may take concerning the heat energy received will tend to be overwhelmed by the high volume of hot airflow. It’s hard for me to put into words completely, but I tended to see darker WB (whole bean) colors relative to GR (ground) colors when roasting on the IKAWA vs. a drum roaster because of that higher airflow. If I wanted to achieve a color spread similar to what I would expect of a drum roaster (while using inlet profiling), I had to use the minimum possible airflow after the first crack to keep the beans moving. Often, this is 60%, which represents the lowest possible setting for airflow on an IKAWA. A huge thank-you to Andrew and the IKAWA team for humoring me… when I turned it down even lower than 60% (with their help), I didn’t see any real benefit or return with the profiles I was using at the time, so though it still seems high… 60% appears to be just fine.

So, when considering the path and airflow volume through the machine, note that the theoretical bean temperature is situated somewhere between the exhaust and inlet curves. Also, starting with a higher airflow and moving toward a lower airflow setting (principally focusing on keeping the beans lofted) is recommended. I’ve tried roasting with a focus on keeping the absolute minimum airflow that I needed to keep the beans moving during the entirety of the roast… but this approach led to a lack of flexibility bean to bean as each bean’s shape, density, etc., really impacted its loft. I recommend starting around 80%, tapering to ~ 70%, and then down to 60% before the first crack.


When considering building profiles for the IKAWA, I would 100% recommend using inlet and not exhaust for the basis of your roasting. For one, it will not vary as greatly with airflow differences caused by altitude or minor differences in fan speed (essentially, it is more transferable). Also, since it doesn’t have to pass through the bean mass before measurement, it removes much of the lag for the PID and allows for a nicer “following” of the designed curve. With the exhaust curve controlling the roast, there is a high tendency for peaks and valleys with the inlet temperature curve, which causes a bit of a chaotic environment for the beans. Please check out my IKAWA 2.0 lesson on Vimeo on demand if you’d like to learn more about Inlet vs. Exhaust.


It’s no real surprise, then, that I followed many of these concepts when producing the roast profile, I am sharing with you today. I am using Inlet air temperature (since I feel it is superior to exhaust air), starting with higher airflow (80%) and dropping steadily to reach minimum airspeed about 1:00-1:30 minutes before the first crack. I have also planned this profile so I am reaching the maximum inlet temperature at about the same time as the minimum airflow volume. This is similar to how I often roast on a drum roaster (though not intentionally). I generally begin decreasing my power input as the first crack is on the horizon so I can enter it slowly enough to hit an appropriate development time without drifting too dark in roast color. From then on, the power steadily decreases until the roast’s end.


Remember, this profile is for an IKAWA Pro50 v3.

I am using 50 grams of coffee during the roast.

  • If you use less, it will roast faster (and likely darker).

  • If you use more, it will likely roast slower and lighter.

  • Highly recommend trying it with 50 grams 🙂

My test roasting is done primarily at 900 ft of elevation (~274 masl). I strongly believe that altitude should not impact roasting on the inlet, though it will change how the exhaust curve looks (but in my experience, the coffee coming out should be similar).

I have tested this with coffee from various regions with various processing methods and had decent results.

Up to now, I have tried it with:

  • Papua New Guinea

  • Colombia

  • Guatemala

  • Kenya

I want to reiterate something I mentioned earlier in the post and make it bold so it doesn’t go unnoticed. There is no such thing as a one-size-fits-all profile for roasting coffee. Every roast profile will need minor tweaks and adjustments to get the most out of the coffee you’re roasting.


Currently, I am running this profile with minimal involvement. The one major thing I do is listen for the first crack and then make sure I have 1:15 – 1:30 development time afterwords by either choosing to end the roast early by sending it to cool-down or extending the roast slightly by manually (quickly) editing the final point to give it 15 seconds more (depending on when FC hits).

My aim when sample roasting is to get between 65 and 70 WB and 105-110 Ground (ground for espresso) Agtron color. I also aim for 11-13% weight loss percentage. I love the expression that Drew Billups gave when he expressed aiming for a certain % above the initial moisture content of the coffee. Drew recommends WL to be 2-3% above the original moisture content. I tend to aim for 1-2% WL% above the initial moisture content.

I’d love to hear from you in the comments and see how this roast profile worked out for you and the coffees you’re roasting!


If you have any issues getting the coffee to hit the first crack around 7:30 (or at all), try raising the peak inlet temperature (the hottest point of the curve) up by 5 C. This higher peak will make sure the coffee gets the energy it needs.

Rob Hoos 1
Rob Hoos 2
Rob Hoos 3


Reminder: inlet (yellow) profiles do not translate across different size IKAWA models.
Rob created this profile for on an IKAWA ProV3 and will work with ProV2, ProV3, and Pro50 models.
Use Rob’s explanation as instructions for creating a similar profile on a Pro100 or Pro100x.
And as he said, add 5C or a little more at the peak if you’re not hitting 1st crack in a good time.

Rob Hoos 2017 profile

Rob Hoos (2017 Profile)

Rob first shared this profile with us in 2017. He recommended it to us for a washed Kenyan coffee. While it’s not his recommendation anymore, it’s really interesting to try it and see the differences between this and his new one.

Rob Hoos 2023 profile

Rob Hoos – 50g  (2023)

This is Rob’s 2023 recommended roast profile. It’s for 50g batches on our 50g models (ProV2, ProV3, Pro50). That’s important to note, because it’s also an Inlet profile (shows as yellow). Inlet profiles don’t share across models (it won’t work with a Pro100 or Pro100x), and they don’t adapt to different batch sizes. However, it’s excellent, so give it a try and compare to the 2017 one!

Follow @IKAWACoffee on Instagram for more.
Read the original post on Rob’s site here.

author-img By Geoff Woodley