Helpdesk Support Required Please - Power Profile Zone Test

[Geronimo]

Hi Kaffelogic Support

Was mucking around trying a power profile zone. Would appreciate someone who can throw some light on why the black power line is not a straight line with a 7.1 percent gradient. The operative word being "straight".

I have used zone 3 for the zone, and not sure if herein lies the problem of not using zone 2? The power zone 3 is from 1'30" to 6'40".

Hopefully it will be blaringly obvious to someone. Log included for scrutiny purposes.

I could have used any profile for testing, but in this case used Steve's MD3 profile for testing purposes.

As I mentioned in opening, Im just mucking around, and intrigued to see what happens using a power profile zone.

Thanks

Power Profile Zone test.png (66.17 KiB) Viewed 6270 times

Power Profile Zone test.klog

(66.74 KiB) Downloaded 356 times

[fnq]

Hi

I have no idea what power zones are supposed to do ( must be further along in the video tutorials that i haven't got to yet) , but keen to learn as well, and maybe in my ramblings I can contribute also.

My guess is the power zone info is taking some priority over the PID function. I only say this , as in my experience the PID programming/sensing has always done a remarkably good job on my logs when i come in from the opening heater sequence too hot and strong.. Ie it pulls me down to the profile curve.

My power curves mostly look wiggly like yours ( part PID on off switching and maybe part temp sensing being a bit erratic in the airflow). I trust the profile NOT the power line. The only time i get too concerned with the power line is if I max out and flatline when it needs to keep rising ( the warnings on KL tell me not to try but sometimes i ignore them at my own peril),..... in these types of low power times i drop below the profile towards the end of the roast.

If you applied a smoothing function to your power line data , like they( KL) do on the ROR(green line) , i think you would be getting very close to your linear goal.

Even on Steve's MD3 profile ( which i mentally envision as an uptick profile), as he is trying to pack a lot of development in a short period at the end of the roast, he achieves this without applying any zone info.

So , again in summary , I have no idea what power zones are supposed to do!!!!! but i do like your willingness to play.

Where did you get the 7.1 percent gradient figure from,. ( measured off the MD 3 profile? , taken from a coffee roaster's secret notes? Cheers Darryl

[Geronimo]

Hi Darryl

Have a read in the roasters companion updated document that includes power profiling. Link at end.

I initially roasted some Brazil Cerrado using MD3, and as I needed a second batch, I thought I would muck around and apply a power zone, purely from an experimental point of view. The 7.1 percent comes from this initial roast using the MD3 profile. See companion document for the method.

The power curve when PID is disabled should be linear. After I saw that it wasn't a straight line (no wiggle), I looked at some of Steve's power curves and they are perfectly straight where the power zone is applied. Thus the initial posting on asking what have I missed in getting this to work.

It is definitely better to utilise PID control due to variations from roast to roast. If using power zones, they are bean specific, and even then they might not work as desired (manual talks about why), in particular when the zone starts.

The initial roast wasn't too bad, and I could have used another profile to test, but was keen to try the MD3 iteration. Considering I like to let the beans sit for 7 to 10 days before they are any good.

https://kaffelogic.com/downloads/manual ... on%203.pdf

Cheers

[fnq]

thanks for the link ( i hadn't spotted it.) Cheers D

[Steve]

Simple mistake!

Good you are trying to understand what is actually going on with power profiles which ARE more important than what the actual roast profile does. This was the whole point of my recent thread, if the power profile / black power line shows large reductions in power or long periods of flat line then the roast will be underdeveloped and or baking PERIOD and there will be varying levels of defects. The reason this is true is because for this tiny little roaster to ROAST it needs to be continually increasing POWER all the way to the end of roast otherwise the roasting environment temperature will drop and this creates roast defects. This is most important after the coffee has gone through yellow and started to brown / FCS and beyond.

Blue arrow is my "starting power profile" Zone 1, which was the whole point of my recent profiles thread = starting the roast with sufficient / steady power = good development.

Green arrow Zone 2 is where you should have input your power profile setting which looks pretty good to me for a first attempt.

Red circle = Zone 3 where you could try putting another power profile after your first one. Say just after first crack you wanted to ramp the power steady all the way to 2nd crack and this would pretty much guarantee your roast will not be baked by the PID being overly sensitive if you can get the % increase right.

As far as I know (besides manual mode) power profiles are the only way to stop the PID from reducing power during the roast. If the PID could be tuned to NEVER reduce power during a roast this would be great for everyones coffee! As it stands as soon as the average temp reading (yes that's right its just an average of the probes readings) goes slightly above the set profile line the PID reduces power and every time this happens during the roast it is creating roast defects because there is no thermal mass in this roaster to fall back on its just a stream of hot air (like me).

Gpowerzones.JPG (149.83 KiB) Viewed 6226 times

[Geronimo]

Many thanks Steve

Very well explained, and makes so much sense, and grateful for correcting my rookie mistake. At least it’s an easy fix.

Looking forward to further experimenting.

Cheers

[Steve]

No worries.

Inspired me to go back and look at some of my old power profiles and try something new based on current thinking and setup.

This seemed to work pretty good and its very simple, I think my other PID profiles might be a bit baked compared to fixed power! I forgot how good the fixed power roasts smell, I think I only stopped because I could not control the end temp...

1200W starting / pre heat power
Zone 1 = 0:00 - 2:30
Zone 2 = 2:30 - 8:00 (12.5% boost)
FCS around 4:45 which gives plenty of time for post first crack temp increase / darker DEV.

Stop roast manually as you see fit and worry about what the average temps did after tasting the roast.

[Steve]

Here is a good example of an entire roast running on a power profile up to Starting Second Crack.
2:30 / 2:30 / 3:10

Starting power 1150W

The Teflon wrap on my probe has burnt out and needs replacing, which goes to show bean probe readings / pretty profile lines on a screen really do not mean much. Pretty much guaranteed this roast is not baked, is it optimal speed for the beans I used, who knows?

The Fan Profile used was from CDFP roast profiles.

SSCPower.JPG (86.69 KiB) Viewed 6212 times

PowerProSSC.JPG (28.31 KiB) Viewed 6212 times

[Geronimo]

Steve it looks like an earthquake struck during your roast. It will be interesting to see how it tastes.

Cheers