Translating an IKAWA Profile to Production Roasting
How do I go about translating an IKAWA profile to production roasting?
This is a question we get asked a lot. And when you have nailed your approach, the time and cost saving that you can realise are substantial. It’s much faster to do several profile roasts on an IKAWA than your production roaster, you use way less coffee (which is expensive!), and can easily try more profiles roasts, allowing you to optimize your production roast and really make the coffee shine.
In truth, there is no magic bullet. There is no formula that simply allows you to replicate your exact IKAWA results on your production roaster – but there are ways you can use your IKAWA and knowledge of your production roaster to get very close.
In this post, we speak with four coffee professionals who profile on an IKAWA and use the insights from this to inform how they approach their production roasts.
Before we jump in, it’s important to align on terminology. We’ve previously talked about the difference between sample evaluation and profiling; IKAWA roasters are excellent for both – if you haven’t already, have a quick read of this blog to align on terminology. In short, while IKAWA is ideal for both sample roasting for evaluation and profiling, the purpose and goals are very different: Sample roasting is to evaluate the qualities of the coffee whereas profiling is to explore the potential of a coffee – which can be really useful to help determine how to approach your production roast.
Below is the general approach we recommend – and as you try this more and more, you’ll be able to hone the way you go about it to meet your particular goals.
Translating an IKAWA Profile
1. Use your IKAWA to explore the potential of the coffee you have, and how different roasts achieve different flavours.
2. Once you’ve found your preferred roast on an IKAWA – consider, how does this roast compare to your ‘standard’ roast? How does the colour compare – is it lighter or darker? What about development time or development time ratio (DTR) – for how long did you roast the coffee after first crack – compared to your standard roast?
3. With this in mind, consider your standard production roast curve – and apply these parameters to it. Simply – if your preferred profile roast on IKAWA had a shorter than average development time and was lighter than your standard IKAWA roast, design your production roast to have less development time too.
While this may sound very simple, it does rely on you having a good ‘standard’ production roast to start with, and confidence in your sensory evaluation of the profiling you do on your IKAWA – after all, cupping the same coffee roasted with small differences can make it hard to perceive differences in flavour.
IKAWA Cup is a great place to capture your cupping notes, from both IKAWA roasts and ‘External Roasts’ – read more here.
Below are some of the ways that experienced IKAWA users have fine tuned their approach and advice. The approach above, and those shared by Morten and Emily are applicable across all different types of production roaster, including drum roasters like Probat, Dietrich, Giesen amongst others, as well as Loring – which we have a specific example of below.
Morten Münchow from CoffeeMind takes this approach further.
His research has found that fundamentally, colour and development time are responsible for 95% of the flavour variation in a coffee – so these are the key areas to focus on.
‘It is possible to copy methods, but not results. It is important to be consistent with the taste and colour between your IKAWA roast and production roasts.’
> What is the roast degree you should achieve for the brew method and audience?
> Therefore, what colour should your production roast be?
> And what is the optimal development time?
Use your IKAWA to help determine this – you can experiment with small batches at low cost. With this information you should know your ideal colour and development time for your customers. This is the basis for your production roasting – but it will naturally vary a little for different coffees.
Before jumping in, Morten recommends some benchmarking work based on colour mapping; create a Standard profile on your production roaster (not based on what tastes best, but on some other specific parameters) – then create a Standard roast profile on your IKAWA that achieves the same colour. This will get your flavours within the same ballpark.
Then roast the coffee on your IKAWA with varying development times – all achieving the same colour that you have determined is ideal for your customers.
Cup the coffees, and take the roast that tastes best. Note the development time, compared to your Standard IKAWA profile, and apply this change to your Standard production roast. Morten believes this approach will achieve translation from your IKAWA to your production roast with 95% precision.
Emily Jackson is IKAWA’s Coffee and Content specialist and uses IKAWA during her day to day work. She has gathered years of production roasting experience while working in the US and during her time at London-based roastery, Climpson & Sons.
Emily’s approach is similar to the ones outlined above – but has a simple framework matching key moments in the roast based on percentage of total roast time.
When it comes to Production Roasting, Emily tends to have a standard shaped curve that she’ll use for pretty much all roasts, but will change the curve slightly to hit key moments in the roast – ie colour change / yellowing and first crack at specific times – so her main goal in profiling is to determine what these times should be.
Emily likes to use IKAWA Sample Roast 2 (which you can view in our Online Roast Profile Library) as the basis for her profiling. She will mark the time for colour change, and first crack, and pay particular attention to the DTR.
She will make small edits to IKAWA Sample Roast 2, modulating the time before first crack, and development time after, seeking to make the coffee tastes as good as possible – and in her experience that tends to happen when the DTR is 13-15%. This usually takes 3-4 roasts. All her roasts are within quite a narrow range.
Next, she blind-cups the roasts, and checks which IKAWA roast achieved the best results.
Using the Roast Log, she can see what proportion of the roast time was in Drying phase before colour change, what proportion was during Maillard between colour change and First, and what the DTR was – and apply these percentages to her planned production roast.
Finding the best production roast is often about quite small margins. If you’ve got your new season’s Kenyan AA and plan to roast that, naturally you’ll reference the production roast you had for last season’s Kenya AA – and use that as a start point. But there is no guarantee that this same curve will taste excellent – modulating the development time by as little 20 seconds may have a substantial effect – so finding that perfect development time on your IKAWA will save a lot of coffee, and really help your production roasts to shine.
Of course, the actual amount of time won’t match up as production roasts tend to be much longer than IKAWA roasts, but the percentages from IKAWA determine the time on production roast curve.
The other thing that won’t match up is the temperature readings. The absolute numbers on the IKAWA will likely be different to your production roaster – due to a multitude of factors which Rob Hoos wrote about here – but this does not matter. The main thing to pay attention to is the shape of the curve itself and timings. Emily uses Rate of Rise (RoR) as a proxy for this and seeks to have a comparable RoR on her production roasts.
Patrick Seeney is the owner of New Order Coffee Roasters in Michigan, USA, a small-batch roastery and cafe business launched in 2017. Patrick switched to using an IKAWA Pro after using a traditional sample roaster for a few years. He enjoys the consistency from roast to roast, values shareable profiles and the ability to create and store profiles on his phone.
Patrick believes that automation is the future of coffee roasting and that his IKAWA Pro allows him to actualise that future. His production roaster is a Javamaster 1500, which has a high degree of automation. In a difference to Morten and Emily, Patrick seeks to find the ideal roast on his IKAWA, then map it across in a relatively like for like way to his Javamaster: ‘The Javamaster allows for editing of 7 segments of the profile, which correspond to the temperature points that I like to hit at specific time points on my IKAWA profiles’.
Green coffee volume and turnaround times on an IKAWA tend to be very different to larger capacity roasters. This difference has been one of the most challenging steps Patrick Seeney has experienced. “In the IKAWA Pro, there is no drop in temperature when you drop the beans into the chamber as you set a start temperature in the profile and the machine follows it”, he explains. “In any production roaster, the temperature will inevitably drop as the beans enter the chamber, until turning point, with the beans getting momentum and the temperature rising.”
In order to solve this issue, Patrick has done some ‘reverse engineering’ and added an extra step to his IKAWA Pro profiles, mimicking the temperature drop at the beginning of the roast. He believes this helps to translate more easily the different temperature changes, and the shape of the exhaust curve on the IKAWA Pro, to his production roaster.
Due to the capacity and heat transfer mechanism on his Javamaster, Patrick is able to even design roast profiles where actual times match up: ‘Often, the roast times on the Javamaster and the IKAWA are the same, which allows me an opportunity to taste the coffee roasted on the IKAWA before moving into production roasting.’
He uses the inlet curve on the app, something less common amongst IKAWA users. “With inlet curves, the roaster is always applying the energy you prescribe, but the coffee is allowed to respond naturally to that. This opens a lot of very interesting possibilities”, he explains.
Alex Wallace started developing inlet curve profiles to look at how fast heat is being applied to a coffee and how fast the coffee will take on that heat. He also seeks to calibrate the colour range and type of extraction from the IKAWA roasted samples, with what they are seeking on production roasts. “When those profiles are stabilised and you don’t change them for a while, you start to get an understanding of the coffee. Is it cracking really early or not? Are the beans roasting smoothly? Is it behaving quite differently from similar coffees (same origin, density, etc)?”, Alex describes. This data can be very helpful when moving onto the production roaster.
Alex also recommends matching your IKAWA airflow settings to the type of production roaster you have. “If you have a production roaster that has airflow control and you use those settings, that’s where you could start playing around with fan speed on your IKAWA. Because you are using a small quantity of coffee, it could be a way of seeing if the coffees will respond well to the variation in airflow on your production roaster.”, he says. “If you have a very fixed airflow on your roaster, then have the most simple fan speed settings on your IKAWA roast. It can’t be a perfect 1 to 1 match but try to avoid making it too widely different.”
Inlet Profiles for 50g roasters do not translate exactly to a 100g roaster. The Inlet Profiles featured here are for 50g roasters. They will work well on IKAWA ProV3 and Pro50 roasters, but will need to be adjusted for Pro100.
Exhaust profiles are compatible across all IKAWA Pro roasters.
Learn more about how IKAWA can help your business.
We hope you’ve found this article helpful – to talk more about the different ways IKAWA sample roasters can be valuable in a roastery, share some simple frameworks and actual examples from some IKAWA users.
If you have any questions or feedback on this article, please feel free to share it with us – email firstname.lastname@example.org and we’d love to hear from you.