Mimicking the Pro


I tried an experiment yesterday using Burundi Muninya Hill greens. The idea came from Geoff’s comment that we could try to hit the same 1C time as typical Pro profile.
I’ve been reading the Royal coffee blogs (see resources) and they do an analysis on the greens they roast using a Pro Ikawa. Typically they hit 1C around 4mins. So I gave it a go. Here is the profile

‘Marshal etheo’ freshly roasted by @ikawacoffee, here’s the recipe #IKAWAHOME - https://share.ikawa.support/profile_home/?CAESEKUzzzqHW0MCg3GMH6PhYyQaDU1hcnNoYWwgZXRoZW8iBQgAEPQDIgYIhQcQ2g0iBgjuDxCEEyIGCJgcENQWIgYIhSMQzxYqBQgAENABKgYIhSMQvwEwAToGCLYqEMcB

I put the beans in the loader but did not drop them. Started the profile and at 1:30 turned the loader to drop the beans then resumed the roast. I got a “doser open” message on the app that stayed for the rest of the roast.
So the beans got dropped into a 175C charged roaster. I noticed yellowing at approx 3 min. 1C came on at 4.30 at about 265C and continued to 5:20. Stopped at 5:30. Moisture loss 12.84%
So a 4 min roast. Missed my target by 1 min too soon.
Now the beans looked good. Nice even roast, no scorching. Opened them this morning 22 hours post roast. Surprisingly nice aroma. Brewed up in the Bonavita immersion Brewer and it was really good. Amazing sweetness, brown sugar, notes of tamarind and the grapefruit acidity. Honestly struggling to tell the difference between this and the Ikawa Home profile.

Got a bit of a backlog of beans to drink now but I’ll be returning to this type of experiment for other beans in future. Seems like I got lucky!

What have I done wrong?

Glad you liked my profile!!! :smiley:

I’ve found no difference in trying to “charge” the charge (dropping cool beans aka charge into a “charged” roaster. Because the thermal capacity (heat retention etc) is quite low. However a prewarm up does help a lot for my profiles. I do a constant 490 for 8 minutes but I stop it short if I’ve sorted and my beans faster then the length of the prewarm-up.

Also nice notes on the roast. It I suggest not flattening out on the profile. Something about a flat doesn’t work the way it’s imagined to…my guess is that the residual heat capacity of the coils keeps the thermometer hot but as the air expands through the chamber it is losing a little bit of heat. The pro moves like 5*F the whole dev phase, and Jens profile is like 7mins. Why are you trying for a 5min profile? The profiles I make require a couples days rest though drinkable at first the real flavor blooms in time.

Lastly I have decided that what is in the bean is already there. A profile can express flavors differently (more or less), and effect optimal rest time. Maybe help find something that wasn’t by there before. But beware that beans can move on you so when a target has been placed doesn’t mean since the profile hits it the flavors will come out…profiles need to be adjust according to age.

On further inspection that isn’t not my profile, my apologies. The Marshal Etheo title and look of the curve threw me. I’m glad though that you’ve tried a more gentile approach for the dry ramp (1st segment). Also a WARNING: that profile goes over the recommend temp/duration specifications


what a novel idea! Your clearly not supposed to do it that way, but its a great way to mimick a pro cycle (well in Theory at least).
I fear though that there are more differences between the consumer and the pro version which potentially can have a (great) impact on the roast. Still interesting idea!


I just so happened to have run your “Marshall” profile on my last roast that it was the first screen that came up when I turned the app on, so I edited it and clicked “overwrite” rather than “save as copy” which would have been more sensible. Apologies for ruining it😀
As to why 5 mins? I was having a stab in the dark at getting to 1C at 4min as in the Burundi that is analysed on the blog. I’ll check it again but pretty sure it’s approximately 5 min roast.
Next time I will lower the profile to try and get closer.
The manual says that it is safe to stay in the 275-290C zone for 2 mins and 250-275 for 8 mins, so I think it is pretty safe. I was only in the upper zone for 30 sec at the point I stopped it.

I was really surprised it turned out ok on such a short time. Food for thought indeed.

Just about every Ikawa roast i’ve done so far moves day by day, so I expect it to have changed by tomorrow.


On the Pro you can select a preheat temperature and it looks like it takes about 1 min to get there then it tells you to drop the beans, so I’m just mimicking that. I wonder why the Home doesn’t do this? Maybe Geoff can answer?

I guess it uses up one of the five points though. But those Pro profiles look incredibly simple. So for say a Brazilian you can set it lower.

I think the Home can potentially do a good job of mimicking the Pro but it is finding out how to recreate those straight lines that is the challenge. The Pro is automatically adjusting the inlet temperature to keep the exhaust profile straight and uses a more sophisticated sensor setup to do it.

Worth a few experiments I think. Of course it might lead to nothing particularly worth while.


you could consider spiking it to 290 in one minute to compensate for the heat retention when the beans are dropped, then drop and start a ‘normal’ profile from thereon. So your points could be:
I’ll give it shot this evening, I like the idea


Yes, I think the same. But my question is: where is the turn around point that we are looking for? I think I would set the bottom of that spike (160*C in your example) to be that grabbing point. But with the limitations that we have I have tried to steer (failing 5lbs of Guatemalan that I ended up finishing in the oven for that coffee flavor) away from that until I can get more comfortable with being able to call when yellow, 1C etc (getting familiar with this roaster)

Oh my ego…lol jk, no worries. I am glad you are looking for short roasts because I am looking at longer roasts and hopefully when we delve into the other side we will have some solid advice to give.


If I monitor the exhaust temperature next time, I should be able to detect the fall and rise and get a good estimate for the turning point with the thermapen.
My guess at the moment is 200-205C. I added the beans at 175C. Yellow was noticable by 230C.

I have a couple of questions about your Marshal profile but I will put them on the fan speed thread. I think that was where I got it.


First attempt was an utter failure:

1C never happened


2nd attemptook better:
Drop @ 1m
Yellow @ 3m (aka start of maillard)
orange @ 3m30
brown @ 4m20
Chaff @ 6m
1C @ 7m
LC @ 8m10
cooling fase @ 10m10

coffee by the way is a Brazilian (JR Farm).
@Geoff_IKAWA: the pics are huge. is there easy way to make them smaller on the forum?


probably by not compressing the image when you are sending it from your photos to you email.

Looks like you got beans finishing at different ending temps.


and to see whether there’s actually a point to this exorcise I’ve tried to duplicate the result without the pre-heat and drop. Interesting results. Basically I moved all points 1m10 to the left creating this profile and upped the 2nd point a bit.
Drop 0m
Yellow 1m20
orange 2m
brown 2m50
1C 6m
LC 7m10
cooling 9m10.

I let them rest now (as I go for espresso and its late) and will make two cups tomorrow.

edit: @deven.patel411 is there a quick way to compress them? To be clear I use discourse on android to upload them directly. Screenshot and photo.

you mean due to the different colors they have? I took a closer look and most if chaff and lighting you see, but there is a bit of color difference between the beans. Perhaps I shoul up the fan.


Maybe it’s not that then. I usually put a space between the links. I’ve never had that kind of problem but it did take me a hundred tries to first upload a profile photo. I can get it now by the 50th…movin my way on up :smile:

I meant the variation of the color on each bean and as a group. I mean it’s really hard to judge a photo and all that noise but as long as it tastes good that the point. How I taste for it is by the beans having a certain acrid taste right at the finish also there isn’t a lasting flavor that resonates on the palate.

Uuummmm That I am not sure. If it is similar to what I’m doing but just compressed. Then the fan high in the beginning will help even the dry but for big light beans I like a big dry because it takes a long time to really dry. 8 minutes isn’t exactly a short profile and i would defer to my Marshal Etheo Shakiso Natural Sidamo Guji, Ethiopia

An thought just came to me, if instead at look at the Inlet Temp on the graph as the actual inlet temp but instead looked at as more of a graph like how gas roasters look at how high or low the flame is and dialing in the right amount of heat. Than for each segment we can look at it in 90* segments and base inlet ROR as % of power of that segment. So I charge at 45% power, than reduce it to 35%, 18% an 10% power. (for Marshal Etheo), below is A profile variation.

In addition

this way of looking at the profile may help smooth out the profile. Maybe I should explain the circle one better. If looking at the total potential energy that the system can produce as once circle from beginning to end, as time (y axis) and temp (x axis). Then we can assume that is the total energy possible. When breaking up the segments is like breaking that one big circle of energy into as many circles as segments. Now if each segment contains a portion of that total energy we can adjust a segment proportional to how much can be added or subtracted from the other circles. In order to optimize the energy of that curve than the Inlet temperature as a line should be the diameter of each circle.

Method to convert pro to home profiles

edited my former post and also put a space in between, but not much happening. Note that if you click on the image you get the full size.


I’ll have to read deven’s topic a couple of times to formulate an answer.

In the mean time: SHOWDOWN (between the 2 roast methods, posted hereabove “pro sim 2” and “3”).

I’ve waited until the next morning to do a tasting because I taste the coffee as espresso and pulling an espresso from a fresh coffee has its challenges. Biggest challenge is the fact that there is quite a build up of CO2 in the coffee that, which during pulling a shot pushes back making it harder to get the coffee through the puck. For this reason I need to adjust the grinder to a coarser level so coffee comes through at all and the time to pull the shot is within the limit (25-35s). I’ve managed tis quite well on the first try (I got 40ml in 30s out of 18g of coffee, which is SCAE target, personally I like 60ml better) however

  • CO2 contributes to acidity
  • shorter development contributes to acidity
  • coarser grind contributes to acidity
  • underextraction (so 40 instead of 60) contriubtes to acidity
    meant that I was comparing two acid baths in tasting…

Was there a difference? Yes, I found a slight hint of caramel in the pro sim 3 profile, where this was completely absent in the sim 2. Given that I pulled the shots under nearly equal conditions and with the same (observable) result, this effectively means the the pro sim 3 method has developed the roast slightly more.

This evening I’ll try to ‘over-extract’ the roast to see if I can diminish the acidity somewhat and get some other taste out of it then coffee vinegar.


so last evening pulled a shot with ‘everything wrong’ (96C 40s, 60ml) to see if I got a better result: the acidity was (nearly) gone, but what remained was bland and watery. So this morning I tried a different approach (high temp, finer grind, more coffee and longer pull), but my equipment wasn’t at full temp yet and I had to go so I pulled a shot at ~90C 40s and 30ml…acid bath again.

Overall I find this the biggest challenge in getting your roast right…its not the only variable. So far I’ve noticed far (really really far) more difference in pulling a shot with a few seconds off, than the difference in a roast with a minute off.


Rolph is there any reason you don’t try an immersion brew? Maybe you just prefer the espresso shots but it sounds like it’s getting in the way of evaluating the beans until they have degassed a bit more.

I don’t cup in the conventional sense as I’m going to brew it anyway. I know a few pro baristas that use the Clever brewer rather than the conventional cupping protocols.


yes, convenience… I nearly only drink espresso, have the machine at hand and is the fastest way for me to get a shot.

But your right, the though has occurred to me that perhaps I should use a cupping method to better test my roasts, alternatively I do have and aeropress lying around and a haro v60 which both are a lot more forgiven in the brew method.
however in the end I have two issues with that:

  • hassle (granted sometimes I like a V60 or Aeropress and then I use it but 80% of the time I go for espresso)
  • I want a good roast for espresso. Adding yet another layer in between will make it that much harder to get the right coffee I like.

but thank you for the suggestion!


Presumably you normally let the beans rest for a few days before trying as espresso?
It’s only if you want to evaluate them much sooner that creates the problem. It might be a waste of time evaluating them sooner as they might not have developed the qualities you like so soon?

But if you were to use a brew method then simplest is best so as to exclude as many variables as possible. I wouldn’t use pourover as you have a big variable in the pour. Aeropress has the plunge. Personally, I don’t much like the AP. French Press is too silty for me. Hence I use the Bonavita (very similar to the Clever). I always use the same brew weights coffee/water, same grind, same brew time. It’s very simple. But very consistent.


Hey @rsegers , what machine are you pulling your shots with ? I would say 96C is crazy high, and its not that simple relation with temperature either (today I persuaded a barista to change the temperature from 93,5 to 91,5 on a natural ethiopia that was coming out kind of flat, with not much body and not much fruitiness and acidity, and it made huge difference in the cup - more fruits, better acidity, more body overall)
I do most of my tasting maybe 10h postroast, and never go above 93C (usually 91-92) and dont have to grind coarser either (finer maybe, but usually not).
CO2 is a big factor though - did you try the Nitro flush? Its great and you can pull espresso right after dropping :wink: