OK, here's the preamble: For some time, my home roasts haven't been great from start to finish. I usually open after 10 days rest and 5 days later I more often than not discard the remainder (50-80g) because it loses its flavour.
I recently bought a bag of CS Espresso Wow and it has been excellent from opening on Day 6 (post-roast) up to now, Day 19. While I'm never going to achieve the same quality as Andy's commercial roasts, I wonder what are the main factor/s affecting how long a roast remains palatable. Roast profile? Type of roaster? Storage (I only use 1-way valve zip lock bags)? Bean type? Bean age (I have several greens which exceed the 3 year 'limit')?
So my question is how do I make my roasts last longer? Occasionally, I do get one that stays good to the last drop but not sure why.
Not sure I can be of much help here Flynn.
Originally Posted by flynnaus
Couple of points, I roast batches of 750 grams green, I usually start using them 3 days post roast, a batch will last 7 to 10 days and usually continues to improve down to the last dose.
I also store in one way valve bags, cool kitchen.
Some of my green stash is getting on in age, probably past 2 years.
I use a variety of beans from various origins, have a preference for African and South American, don't use beans from SE Asia, I dislike the earthy overtones.
I roast quite dark, about CS9, use a Coretto, start from cold, approx 14 minutes to FC, approx 19 minutes to end, just before second crack.
Don't use roasting software, monitor bean temp and time with a DMM and watch.
Cool rapidly at end of roast and bag immediately.
What roasting method are you using?
Everything else being equal, the roast profile will have the most significant effect, given that Andy's Wow is going the distance. Have found that the deeper one roasts, the shorter is the time for optimum and enjoyable consumption and in the case of dark espresso style roasts, these seem to better if opened earlier rather than wait too long. Different rules apply for Yemen Ismaili though... ;)
As Yelta mentions, knowing how you roast your batches will help us to know what might be happening and even better if you have a sample of Roast Profiles that also could be checked out. :)
Thereís really only three possible causes:
1. Storage - you seem to have that sorted though.
2. Roast depth - if youíre roasting quite dark this could happen. For a standard Ďespressoí roast Iíd suggest 3 days rest and 10-14 days of use. If youíre roasting lighter for filter or manual espresso then the most likely problem is......
3. Green bean age - there are many factors that will influence this (green quality, processing, etc.) but generally once you reach the 12 month mark after harvest youíll see fairly obvious deterioration unless youíre storing them in the freezer.
I use a KKTO and my roasts are stopped at the start of or just before second crack. I have to use roast monitoring software as a combination of an enclosed roast chamber and poor hearing means I can't hear the cracks. I'll try to upload a profile but my basic technique is to
- Preheat to 170 deg
- Add the beans (450 gm of SO or pre-blended)
- Maintain a RoR of 10 deg/min until 140 deg
- Drop back to 6-7 deg until 155 (for Maillard reaction)
- Turn up and continue at 10deg/min to 195 deg
- Drop back to 4-5 deg min until it reaches 220. Roast time is about 17-18 mins
- Dump the roast in my bean cooler and stir while cooling for about a minute.
- I split the roast into two 250g bags to try to keep one half fresh
I wondered if my KKTO was the problem but had a similar longevity problem when I used a Behmor while my roasting lappy was out of action/ This is what led me to believe bean age might be the reason, reinforced by some of my older stock going flat in about 4-5 days. My next roast will be of newer beans to see if I can get better longevity.
I guess, by my own standards, those roast rates before and after the Maillard phase slow-down would be considered quite slow.
Probably the slowest I ever use would 11.0C/Min but more usually, up around the 13-14.0C/Min. I don't pre-heat the BM Pan though, prefer to start from ambient temperature as pre-heating never realised any improvement in the cup and it doesn't really replicate what's going on in a commercial drum roaster - Much more Thermal Inertia to consider with them and the roaster must be pre-heated to a normal operating temperature in order to achieve any resemblance of an acceptable roast profile.
What used to work best for me when I tried the KKTO, was to incorporate a 'drying phase' with the beans in the roaster and allowed to heat gradually to approx. 120.0C bean temperature and held there for approx. 3-4 minutes before starting the roast profile proper. Resulting roasts were always very sweet with no obvious flavour deterioration until the bag(s) were finished. Sure, the flavour profile does change over that time but always a very nice brew resulted.
We keep our green stash stored in a cool location in the centre of the house at the bottom of a dark cupboard. Have successfully roasted beans that were two years+ from purchase date and observed no 'off tastes' in brews resulting from them. In fact, pretty well every roast batch was smack on the button of what I expected, and I'm pretty fussy. Sure, fresh is always best but there's no need to freeze the beans if you have a decent storage location set up in your house somewhere.
Don't know if any of this useful to you Steve but maybe try sharpening up the temperature gradient as described and see what that does for you.
All the best, :)
My KKTO struggles to get those temperature gradients but it still produces tasty roasts, some beautiful. I did try the drying phase early in the history of using my roaster but was still ending up with long roast times. I tried using a metal plate under the inner pot to reduce the roasting chamber volume with some success. Smaller batch size work but somewhat impractical andfound 430g is a happy medium.
It wasn't until I resorted to preheating that I was able to get decent roast times. Yes, I understand it isn't a drum roaster but it still contains a significant quantity of metal.
One of the problems with the KKTO is lag. When I turn up the heat after Maillard phase, it takes at least a minutes for the gradient to climb to 10 deg/min and doesn't rise above that until first crack exothermic kicks off.
I've applied all the KK recommended mods over the years. I still have my old corretto; might be time to dust it off and gain a bit more control over the roast. It's a single pan BM but easily capable of 500g batches which is more than enough for my needs.
Ah, no worries Steve...
Sounds like you are roasting up against the limits of your KKTO setup.
Don't know what's available re: TOs with more heating power, might be worth checking out... :)
Just read your original post and I see your problem - 3 year old green coffee. This is waaaay past itís best if it hasnít been frozen. No reason you canít still roast it and get a reasonable result, you just need to realise that moisture content and water activity will be really low and your expectations need to be realistic. Youíre also roasting on the darker side so Iíd suggest cracking open a bag after 24-48hrs rest at most and youíll get a week out of it all going well.
Halogen ones are the go but hard to find a good one. KK recommended another mod of cutting out the centre of or drilling larger holes in the shield that covers the element which I've never tried.
Originally Posted by Dimal
Here's my roast today of 420g of Zimbabwe Chimanimani (bought in Oct 2017). M=Maillard phase, F=First crack
Nice chocolatey aroma and stopped just shy of second crack. In fact, when I lifted the TO, I heard the single crack that announces second crack is imminent. Pleased with the result. Let's see how long it lasts
Attachment 22062 Attachment 22063
You're probably right Leroy. I've had to toss some old beans last year as they had nothing in them flavour- or aroma-wise..
Originally Posted by LeroyC
Dark is a relative concept. As mentioned, I prefer to pull just shy of second crack. See my most recent roast in my previous post. That could be counted as dark by manual brewing standards. We'll see how it goes when I crack this one in a week's time.
Where does the idea that Maillard reactions only occur in a narrow temperature range come from?
I'm not trying to be fractious, it doesn't make sense to me which leads me to think either there is something I'm missing or some there's some serious misinformation out there.
It doesn't - according to this wikipedia article, it ranges from 140 to 165
Originally Posted by Lyrebird
At 15, I turn up my thermostat because the lag means it will be ~ 165 when max temp is applied.
I think that's a serious misinterpretation. What the article should say is that it requires a minimum temperature of 140 - 165 to get rapid Maillard browning in a cooking process.
Maillard reactions occur at room temperature if you give them long enough. They occur more rapidly as the temperature gets higher until they are so fast that the reactions are substrate limited not temperature limited.
There are lots of ways of describing the change of reaction rate with temperature, Maillard reactions follow Arrhenius kinetics but that confuses most people since it's an exponential function. A simpler approximation is to use the concept of Q10, the rate increase that accompanies a 10 degree rise in temperature*. At coffee roasting temperatures a Q10 of x 1.5 (eg the Maiilard reactions increase by 50% for every 10oC rise in temperature) is a good approximation to the Arrhenius equation with an assumed Activation Energy of 100 kJ / mol (the literature values vary all over the place but they're mostly close to that figure).
* That's still an exponential equation by the way, it's just a way of saying it without using actual exponents.
I haven't been able to find any good information on substrate limiting during coffee roasting. A further complication is that substrate availability will be influenced by moisture level but there's even less info on that.
Originally Posted by Lyrebird
I'm not claiming to be an expert. Discussion on this site had it that the Maillard reaction for coffee occurs during the yellowing phase of roasting which was claimed to be in the range 140 to 160 degrees. I only know that I achieved a better result in the cup if I reduce the temperature gradient in that range so I've stuck with it rightly or wrongly.
I have to admit that when I next roast I won't be thinking about Arrhenius, Q10 or Activation energy. I will be using mostly my sense of sight and smell, Roast Monitor (on an ancient laptop) plus some basic principles learned on this site to produce a roast for my own consumption. My quest in this thread was to try to discover why Andy's beans were good to the end while my own roasts rarely went the distance.
and I reckon this
Originally Posted by flynnaus
was the answer that made most sense to me.
Originally Posted by LeroyC
Seems to be two schools of thought when roasting Flynn, the scientific approach, document every detail and keep a reference library of profiles for each and every bean/roast, and attempt to duplicate in the future.
Or, the seat of yer pants style, as you say "using mostly my sense of sight and smell"
I am more or less in between the two, I do monitor temp and have time based bench marks, all of my roasts seem to work out ok.
My son in law has also been roasting for many years in a Coretto and relies completely on senses, does not monitor temp or worry about time, notes FC and monitors visually from that point, his roast are remarkably consistent and certainly very drinkable.
Perhaps your problem is down to age of beans.:)
The common misconception is that the Maillard reaction starts around that 150deg mark and ends soon after. Itís quite complex and still not fully understood, but I believe the latest school of thought is that it does start around there, but in reality continues until the end of the roast (unless of course youíre doing Vienna roasts).
Thanks for the input all. Some good info has come out of the discussion. Bean age is certainly a factor and this may be confirmed over the next week when I open my the two commercial roasts I bought this month were so much better than mine, I thought I'd better check my roasting technique as well and I certainly have some food for thought.
I can't remember Matt/DBC ever claiming that the region of the profile, that is allowed to slow-down, being the exclusive domain of the Maillard Reaction taking place. The concept was taken from commercial roasters in Seattle Washington whose experience, based on experimentation, revealed that slowing down the profile in this region of the Maillard phase, improved the quality of coffee in the cup.
I have conducted my own experiments in this regard and found it to be true. The results in the cup are much sweeter right down to the last bean in the bag. I don't know "why" it happens, just that experimentation reveals it does happen. I suppose it would be great if we knew all there is to know why re: roasting coffee in certain ways can produce particular results but most of us here are only amateurs when it comes to roasting, no matter how long we've been doing it so the "why", though it matters, is not worth the time and effort to discover given that roasting 300-800g every fortnight is all we're aiming for while trying to reduce the mistakes that cause mishaps in the cup. I don't have the money, time or inclination to worry about it.
I also roast with the same setup as yourself. I have basically butchered the heat shield on the TO. By using some tin snips and cutting all but the outline of the shield, you will greatly increase your roasting capacity. I roast 750 to 800g with ease.
I follow KK's recommendation and heat the roaster for 5 minutes (at least two cycles for it to turn on and off). I then load the beans in. This will give you a load temp of 250 to 280C. I find any temp below 230 deg and the roast will lag. After 90 secs I get a Turning Point temp of between 83 to 92C, with a steady RoR after that. It then tapers off. I burp at around 7 minutes (160deg) to slow the roast down. I also usually burp at first crack, and tip about 3 to 3.5 minutes after first crack. Each bean is different and I experiment on the temp at which I tip. It is always before second crack, with sometimes the odd snap happening.
I now use the heat probe religiously, but also bear in mind the sight and smell.
With old beans, you have a lower moisture level, and hence a shorter drying time. Try a quicker time to 160deg, say no more than 7 or 8 minutes, and see what the result will be. To achieve this you will have to heat up the roaster a bit more. Just watch the onset of first crack, as it will come fast. Smoke and the temp will guide, so you can plan for it.
With age, some beans will deteriorate, it all depends on the type, quality and processing method, as well as storage. I recently roasted some 2 year old beans that were vacuum sealed, until the day before roasting. They were excellent.
If I followed your roast profile, I found that the roast would drag out to over 20 minutes, but by increasing the drop temp ,it became shorter. Each roaster is different and the way you measure temp, so this is a guide only (as you know).
A TP of 60deg is what commercial roasters aim for, but the larger roaster and quantity of beans is different to a KKTO. We have to compensate.
The only way to find out is to try different temps.
As a wise roaster stated elsewhere, the magic starts from when you load the beans, the TP and RoR after that.
That part is the most important, and will reflect on how the roast behaves later on.
Cheers for that Bosco. Reassuring that my roasting method is somewhat similar.
Never would have thought of upping the drop temp that high but with larger batches it makes sense. I'll give it a burl I have some tin snips somewhere.
I cracked a PNG Wahgi + Peru CDS roast today, 9 days post roast and it was excellent . I hope it goes the distance.
While I am yet to try the Zimbabwe roast, I did start a Peru CDS + PNG Wahgi blend (50:50) on April 30 which I roasted on April 21. Still drinking well almost a week later. I am happy to accept that roasting with 3 year old greens is probably the main reason for poor longevity.
I am also mindful of roast depth as Mal referred to but I don't roast dark by espresso standards: usually stopped at just before second crack.
I find green beans are best up to one year, then roasted provided stored with one way valve make for best drinking from 3 days, up to about 20 days.
Thatís a pretty broad generalisation. Green beans can easily be ruined in months if poorly stored. If stored properly in a stable climate that can be good for more than a year. If vac packed and/or frozen they can be good for much longer. And the same goes for roasted coffee. The Ďlighterí the roast the longer it should last, but again storage makes a big difference. Iíve had well roasted light roasts that have been good for up to a couple of months post-roast.
Originally Posted by zcoffee
I've heard reports by some that they have opened a forgotten roast a couple of months after roasting and found it to be excellent.
Yes Iíve had similar experiences. It does seem to fade faster once the bag is opened, but that just means you need to drink it faster.
Originally Posted by flynnaus
Wish I could say the same, have never been in the position to forget a roast.
Originally Posted by flynnaus
About the closest I have come is having gone overseas for a month or so leaving a partly finished bag behind, on return I naturally use them and to be honest they usually stand up pretty well.:)