Steve's Ultimate Profiles 100g - Constantly Declining Fan Profile (Fan Cal 0.91)

[Mark]

After a scolding from Steve for not understanding the how and why of power profiling (joking Steve)

I have had a deep dive into power profiling … and below is my effort using 3 zones. I used Steve’s “Power” as a base with the intention of lengthening the roast time for an espresso style shot. This is the first tentative effort.
Roast was stopped manually for 20% - 25% DTR.

Colombian Volcan Galenas Supremo is the beans I used here.

Haven’t tried it yet … like to wait a few days!

Cheers Mark

[fnq]

Could you throw up your log as well. please

Without the strong arm of the PID to hold my heat gain in check, my first couple of feeble Power efforts were terrible. ( There is a reason these power profiles are not for beginners like me). I tried to see if i could lengthen steve's base profile to accommodate to medium roasts ( so 2.8-3 not 1.4),, i couldn't do it. My constant power tended to superheat the bean mass too quickly once past colour change.... i think i may need to play with the fan profile to try and arrest that problem ( then xmas got in the way and i haven't gone back to trials yet.)

[Mark]

Klog 300 as requested.




Cheers Mark

[fnq]

thanks Mark. That log shows ( to my untrained eyes) , no stalling or surging in the bean temps all the way through, so fingers crossed it tastes as good as it looks.

I am still working through ( another way of saying clueless) at how different the overall power requirements are on these POWER profiles ( your's preheats at 1000w and climbs only to 1160 watts) where other more traditional profiles can max out at 1360watts -- I know the fan profile treatment is different But still!!!)

My take is these POWER profiles have to be individually and meticulously matched for bean and roast size , whereas the others are more likely a one size fits most. - (didn't Steve say something like that about 3 pages ago? It just takes me a bit longer than most)

Well done on the ColumbianVG. Cheers Darryl (edited for spelling)

[Mark]

Continuing to test Steve's Power profile ... here's 2 Klogs with the same Power settings and using 3 power zones using the same beans ... but very different results. I can't figure out why they are so different ... yet have the same settings,
Same beans, same size (100g). Both set to finish at 222c.
Klog 330 finishes at 8.40 minutes and Klog 326 at 9.37 minutes ... both 222c.

Confused!
Cheers Mark

[kaffelogic]

Continuing to test Steve's Power profile ... here's 2 Klogs with the same Power settings and using 3 power zones using the same beans ... but very different results. I can't figure out why they are so different ... yet have the same settings,
Same beans, same size (100g). Both set to finish at 222c.
Klog 330 finishes at 8.40 minutes and Klog 326 at 9.37 minutes ... both 222c.

Confused!
Cheers Mark

Hi Mark

The Nano 7 measures two things at the start of the roast, and adjusts heater duty cycle and fan speed accordingly. They are

• Ambient temperature
• Supply voltage

As far as accuracy goes, ambient temperature depends on factors described in this post: https://kaffelogic.com/community/viewto ... 1861#p1861

Supply voltage accuracy depends on your supply not being too 'soft' i.e. not dropping too much once the heater starts drawing power. This is why using an extension cord can be a problem.

There are also a number of factors that are not measured, among them

• The beans (size, density, variety, moisture content, etc)
• Barometric air pressure
• Humidity

Looking at the start of your two logs you can see a very close match. It is safe to assume that none of the above factors have made a significant difference. The beans you have controlled by using the same beans, and the air pressure and humidity do not usually make a significant difference when you roast in the same indoor place (more discussion of this in this post https://kaffelogic.com/community/viewto ... =278#p1784).

However, if any of these factors change during the roast the Nano 7 doesn't know about the change.

In your case log330 starts with supply voltage 238V, and log326 starts with 245V. As you can see from the close start of the two logs, the roaster has adjusted correctly for this voltage difference.

From an analysis of the motor voltage in the logs I can see that log330 has a supply voltage shift that happens at 00:34 and another shift at 01:15. The first shift brings the voltage up to approx 241V and the second shift brings the voltage up to approx 244V. This change in supply voltage fully explains the temperature difference that emerges after 01:22.

Supply voltage variations will cause problems with power profile zones, and this is what you are seeing.

[Mark]

Thanks for your detailed reply ... much appreciated.
It seems if I wish to continue my experiment with power profile zones I would need to control the supply voltage... having no idea how to accomplish this ... could you recommend a suitable device?

Cheers Mark

[kaffelogic]

Thanks for your detailed reply ... much appreciated.
It seems if I wish to continue my experiment with power profile zones I would need to control the supply voltage... having no idea how to accomplish this ... could you recommend a suitable device?

Cheers Mark

I am aware at least one user who runs his coffee gear off a UPS. You would need at least a 2000VA/1200W rated unit, and that might be pushing it a little with the Kaffelogic capable of drawing 1400W. You might need advice from someone with more specialised knowledge of UPS's, or just an opportunity to try it out and see if it works for the Nano 7. In any case it's a somewhat expensive and bulky solution to the problem.

Other things you can do are to improve your power supply by plugging in to an outlet on the stove if possible. The stove usually has a 30-50A supply from the mains board, as opposed to the 15A supply to a standard plug outlet. This will help reduce the effect of other power uses in the home on the plug voltage. However, this won't be any help if the voltage fluctuations are coming into the house from the grid.

Another thing is to try to avoid the times of day when the voltage varies the most, but this involves some research into what are the factors that cause fluctuations in your area. If it is industrial uses that cause the variations then evening roasting might work out best.

Finally, I can see some potential for building in the ability of the Nano 7 firmware to self adjust during a roast when the voltage fluctuates even in a power profile zone, however this is a development that will take some time to investigate the feasibility and complete. So it won't be of any use to you in the short term.

[Mark]

Thanks once again for your detailed reply ... checked out a few UPS's ... but a bit too $$$$ to justify further experimentation.

Cheers Mark

[Barry O'Speedwagon]

No worries, just reporting my findings as I go...here is some more fodder putting into action what I wrote about earlier.

Aim was to keep a max roast time very close to 8 min, while hitting the temperatures closer to 2nd crack start. Trying to preserve more of the sweet vibrancy in the coffee which have worked so hard to produce but adding that typical darker caramel roast coffee flavour.

2 roasts, one at level 4 and the other pictured at level 4.3, 15.5% and 16% WL respectively.
This was the same Brazil blend, blended with a washed Colombian 50/50.
Mixed both roasts together and cupped with a 10 min rest after grinding,12g / 200g water / 4 min.

I did not get any pronounced off baked dry or wet aromas it was sweet and chocolatey.
Taste was very encouraging, much sweeter, typical chocolatey caramel roast coffee flavours with a lingering after taste that was not bitter at all, quite smooth, nothing offensive. Really looking froward to trying this out as espresso after a few days rest! Cupping suggests its going to be ripper.

I think its worth taking note of the black power profile log, no flat line or dropping off, ramping the power all the way home = NO BAKE = no aspirin like bitterness with tonnes of caramelised sweetness

Same 1300W start 40 seconds.
Corner starting at 50 seconds ending at 1:13
End by time ratio back to the default 0.33333

Be great if people that enjoy sweet med - dark roasts try this and report back.



BlendMD3.JPGBlend2.klogMD3.kpro

So I gave the MD3 a go with the Brazil Pupled Naturals. Definitely the best result I've had from this bean as a strong flat white. I think I dropped the Level to 4.0, but will revert to 4.3 next time I try.

My current experiment is the Brazil using the BrazilDark profile, which I intend to blend with a Costa Rica Miel roasted with the CDFPDark (I have one bag roasted to 5.9...a bit too dark for me, and a second at 5.2 which looks the goods).