What's a light roast? - Page 12

#111: Post by **Denis** » May 23rd, 2019, 8:19 am

Hello dear coffee lovers.

Today I want to bring to your attention a roaster from whom I didn't buy coffee anymore for over one year. Most reasons are objective and for my personal taste and high quality beans request this doesn't suit me. I guess the Nordic roasters treated me to well and have a higher constant quality overall so now is hard to find such level of greatness.

W/o any more chat here it is WRC, mister Gardelli. Taste wise coffee is not bad, not wow, but you can see the "greatness" in the beans sorting, roasting, defects, grading. And this is probably as he is well known, some of lightest roast I ever encounter. I had no problems pulling the espresso shots with this coffee.

#112: Post by **crunchybean** » May 23rd, 2019, 9:47 am

Denis wrote:Hello dear coffee lovers.

Today I want to bring to your attention a roaster from whom I didn't buy coffee anymore for over one year. Most reasons are objective and for my personal taste and high quality beans request this doesn't suit me. I guess the Nordic roasters treated me to well and have a higher constant quality overall so now is hard to find such level of greatness.

It's always a good thing to step outside every now and then and get some fresh air

. As some say "light roast is the right roast", not that I completely agree.

On topic:

I am now determining roast degree by crema color, I find this a better indicator (for me) than whole/ground bean. I do not have color meters and such so I cannot be precise. But I can get a nice general feeling of how light or dark something is by "frothy-ness" of the creama/head and its color.

#113: Post by **Denis** » May 23rd, 2019, 10:12 am

Crema color is an indicator of extraction yield not roast. The more you extract the darker/denser is the color/crema.

My light roasts from Nordics are reddish, this is an indicator of high EY, and I can easily get yellow-lemon and this is fast extraction with low ey and watery shot.

Some grinders tend to do tiger stripes (specially conical grinders), and muddy- spotted crema other are not doing that and you get a uniform - one color crema.

Here are the reddish coffee with high ey I am talking about, these are all from filter/light roasts.

So crema color is a signature from the grinder type and the extraction you choose. Light roasts are harder to extract, so in general people will get lemon color that indicates a lower EY not a light roast, from my perspective that is a bad extraction. As you can see you can get red-brown colors from light roasts.

Here is a coffee made by me on a conical grinder:

But this is not something you can measure, and because its different from grinder to grinder and extraction to extraction it's not a great tool for doing comparisons.

#114: Post by **RobindG** » May 23rd, 2019, 10:19 am

I am now determining roast degree by crema color, I find this a better indicator (for me) than whole/ground bean. I do not have color meters and such so I cannot be precise. But I can get a nice general feeling of how light or dark something is by "frothy-ness" of the creama/head and its color.

That makes more sense than looking at the colour of the beans.

This is an Italian blend, Martella. It does not look very dark, but I tastes dark. Their roast profile is 27 minutes on a relatively very low temperature. So the beans dont get burnt, but they are developed to the max. On the opposite, you can get your beans black in 2 minutes on very high temperature or with a blow torch, but the inside will be underdeveloped.

#115: Post by **crunchybean** » May 23rd, 2019, 11:39 am

Denis wrote:Crema color is an indicator of extraction yield not roast. The more you extract the darker/denser is the color/crema.

My light roasts from Nordics are reddish, this is an indicator of high EY, and I can easily get yellow-lemon and this is fast extraction with low ey and watery shot.

Some grinders tend to do tiger stripes (specially conical grinders), and muddy- spotted crema other are not doing that and you get a uniform - one color crema.

Here are the reddish coffee with high ey I am talking about, these are all from filter/light roasts.

So crema color is a signature from the grinder type and the extraction you choose. Light roasts are harder to extract, so in general people will get lemon color that indicates a lower EY not a light roast, from my perspective that is a bad extraction. As you can see you can get red-brown colors from light roasts.

But this is not something you can measure, and because its different from grinder to grinder and extraction to extraction it's not a great tool for doing comparisons.

I appreciate your response (what I am going to say is offensive, but should not be considered disrespectful) but you are limited by your understandings. Though some parts of what you said are true, they do not all fit together to make a valid Theory.

What is considered "light" vs "developed" pertaining to the use of a Nordic label. This in the general sense the point I was trying to say (but do not good a good job of doing so), we have have a light roast but developing it further than what would be considered light. For instance chocolate is not a flavor considered light but I can develop further and roast light. Altering the scope or focus of the roast to try and pull in flavors for one side of the spectrum to the other. Another instance is getting as you said citrus/acidic notes in a more developed roast.

I also purposefully did not specify for espresso as you rightfully pointed out the concentration and constitution of the medium will impact the refraction of light.

@RobinG, yes exactly. And to further the point of flavor impact. The breakdown of cell structure will also impart the flavor. What I guess can happen is that: once the flavor is formed in the cells from the liquid molecules, breaking off/down cell structure (carbs) can create more body but also (potentially) start masking nuance.

#116: Post by **treq10** » June 1st, 2019, 1:28 am

Too juicy of a thread not to reply

We can get into technical details of light roasting like color grading, crema color, DTR, etc...but there are so many ways to roast light that it's hard to keep count! I live/work in NYC and have access to tons of coffee roasters both local and internationally...and many call their coffees light to medium but they all have a distinct way of approaching their roast profile. It's amazing, actually - you can really tell how their philosophy of roasting light translates into the way the coffees extract/taste.

With all that said, light roast, when done correctly, should highlight all the origin characteristics of the coffee with a minimal display of roast characteristics (sugar caramelization, maillard). Now, sugar caramelization and maillard components will always be a part of roasted coffee that's water soluble, but in a light roast, it's very minimal.

Lots of people like a balance. I find that the majority of coffee drinkers favor somwhere between 50% roast characteristics or more (the classic nutty, chocolatey body/profile). Light roast would keep that somewhere under 30%...meaning very little roasty bittersweetness. Some light roasts are balanced between sweet and sour, while others are unabashedly bright. Some "light roasts" by roasters who like more sugar caramelization are not technically light roasted (meaning too much caramelization/maillard/or in worst cases baked)...but somehow getaway with it depending on the market.

I like to roast light but with just enough sugar caramelization to hold up to milk. It's somewhere between a true light roast and a true medium roast LOL. It's all so whacky in the end to try to come up with narrow definitions for roast/flavor profiles. Just do and drink what you like! There is so much great coffee out there!!

#117: Post by **Almico** » June 1st, 2019, 6:10 am

treq10 wrote:...I like to roast light but with just enough sugar caramelization to hold up to milk...

That's a good point. Maybe the line should be: if milk curdles in the cup, it's definitely a "light" roast. I have had this happen with Kenyas only so far.

#118: Post by **false1001** » June 4th, 2019, 1:07 pm

treq10 wrote:I like to roast light but with just enough sugar caramelization to hold up to milk. It's somewhere between a true light roast and a true medium roast LOL. It's all so whacky in the end to try to come up with narrow definitions for roast/flavor profiles. Just do and drink what you like! There is so much great coffee out there!!

Might be time for a new thread, but I'm going to take that first sentence and run with it...

I've been having a terrible time getting merely drinkable filter roasts with my new Bullet ([troll]Anyone with an open flame and greens can roast for espresso[/troll]). Not only can I for some reason never maintain a constantly declining ROR in shorter roasts, i've found it nearly impossible to get a roast that spends less than ~45% of its time in the Maillard phase which is obliterating the terroir out of any coffee I use. I'm sure this is all due to a lack of skill, but out of frustration I whipped out my old frankenstein HG/BM inspired DIY roaster that was able to churn out exceptional nordic/light roasts with ease. I have already cannibalized the temp logging bits for another project, which was insanely stupid, but I noticed several things immediately now that I have more experience with a different kind of roaster:

1) The "drying phase" (for the sake of conversation we'll call this the time of drop to a lack of green color in the beans) takes up ~60% of the roast time. The roast then speeds through Maillard and FC, almost the exact opposite of my bullet roasts.
2) FC is roughly half the time than in my bullet (45-60 secs vs 90-120 secs)
3) The aromas during Maillard are much more intensely from the terroir, instead of the generic sugary smells I get from the bullet

My conclusions:
1) Convection is far and away the superior way to initiate heat transfer to beans
2) The longer you spend in Maillard the less of the terroir you'll have in the beans at the end of the roast (side stepping light vs medium here)
3) The shorter your FC the easier it is to roast "light"
4) A secondary conclusion from #2 and #3 is you want to have a relatively high RoR during maillard and FC

I'm starting to believe that if you want to roast light you either need to embrace the soak method and have your RoR peak at end of drying (followed by constantly declining RoR, no idea how this would actually be accomplished), or utilize the hot start fast finish method. But who the hell knows, the Bullet has made me realize I'm a terrible roaster who just happened to get lucky with a convection based approach in the past. Now if you'll excuse me I have another 40lbs of coffee to waste while bashing my head against a spinning drum...

#119: Post by **another_jim** » June 4th, 2019, 1:54 pm

Decling ROR indeed! The difference is kind of obvious if you stop using this mind numbing babble, and think basic heat transfer. On your HG/BM, the environmental temperature rises slowly throughout the roast, so the bean temperature profile is far more straight line. In the Bullet, you are using a high ET drop in temperature, and keeping the ET roughly steady. That means the bean temperature profile is more curved, spending longer in the later parts of the roast. Try a lower drop in temperature and more heat so the ET climbs throughout the roast, this will give you the profile you had on the HG/BM.

Oh, plan out your roasts ahead of time, and stop fiddling on the controls to correct every ROR twitch; the more nervously you roast, the worse it will taste. This is the secret reason people get better results on bigger roasters -- they filter out ROR induce roaster twtchiness disorder (RORIRTD)

#120: Post by **Denis** » June 4th, 2019, 2:01 pm

Here you go, got something special for you, watch at 20:11.
And after tasting over 50 speciallty coffee roasteries with drum roasters (probat, giesen, diedrich, etc) and over 6 with aid bed roasters (loring) all I can say there is a huge difference in taste, uniformity of the roast and the most important, extraction values.