It still seems odd - and I know Bill has mentioned this before too - that when setting the brew temperature on the panel, one has to manually factor in the offset.

Since the offset parameter is known and available to the software, fixing this would seem to be a trivial single-line subtraction in the code. LM's stubbornness in not addressing this little annoyance may be more amusing than anything else, but still...

Can't you unlock the PID & set the offset permanently?

If one knows their boiling pt (alt. / baro. pressure), can they adjust by this method of observing flashing at the grouphead:

http://coffeetime.wikidot.com/izzo-alex...t-settings

or is the above method less accurate than a scace?

what are some of the environments where a positive offset is utilized?

Also, does this offset vary during the day with variables of temp, RH, baro pressure, first shot of the day, and after 8 hours of 1 shot per minute, etc.?

The offset is there to account for errors in the system. It may be the same for 80 machines and then change for a few machines. To add this offset in the code would be irresponsible. You absolutely need this for calibration of the machine. Even Fluke meters (and most any respectable measurement device) have this capability.

The offset is not dependent upon the environment. This is only for error and location differential. The environment will affect your actual temperature by increasing or decreasing your heat loss causing your temperature to change. You might be able to compensate for this but it would change throughout the day. Not an easy task.

It is very difficult to measure the temperature of a boiler using the flash at the grouphead. This method is also dependent upon the pressure inside the boiler. You would have to have a pretty accurate pressure gauge hooked directly to the boiler to make an accurate measurement.

Bill

Billc wrote:To add this offset in the code would be irresponsible.

I don't think Anvan meant that the offset value should be hard coded. I think he meant that the offset value set by the factory (or the advanced user) should be added by the firmware automatically so the user isn't required to mentally subtract the offset in order to set the brew temperature. That's by far the strangest and most unfriendly aspect of the GS/3.

For example, if my offset is -4F, and the desired brew temperature is 200F, then I have to set the boiler temperature to 204F. The firmware heats the boiler to 204F, then adds the offset to the temperature reported by the probe (204F) so that the display will show 200F.

This correctly deals with the difference in temperature between the sensor location and the group head, but it's a very unfriendly way to do it. The user should not be required to know the offset value (or that it even exists!)

Instead, the firmware should do the offset adjustment transparently to the user. It should do so by subtracting the offset value from the brew temperature set by the user, and then use the resulting value to set the boiler temperature at the sensor location. It should add the offset to the temperature reported by the probe and show that value on the display. That way, when the boiler reaches the desired temperature, the brew temperature set by the user and the temperature shown on the display will be the same. The user won't have to know that the offset value exists.

For example, let's say the offset is -4F and the desired brew temperature is 200F. The user intuitively enters 200F in the brew temperature menu. The firmware subtracts the offset value from the desired brew temperature, which results in the value 204F. It then applies heat to the boiler until the probe reports 204F. It then adds the offset value to the probe temperature and displays the result -- 200F, which is the target brew temperature. Again, the offset manipulation is completely transparent to the user.

How, then would the offset be calibrated? I can think of a couple of ways to do it, but the simplest approach would be to use a Scace or similar device to determine the difference between the temperature shown on the display and the temperature at the group head. For example, the factory would begin by setting the offset to zero and the brew temperature to 200F. After the display temperature reaches that value, the temperature of the water at the group head would be measured. The brew temperature setting would be subtracted from the measured value to obtain the offset value.

For example, on my machine, when the offset is set to zero and the brew temperature is set to 200F, the display temperature would be 200F (per the algorithm above) and the group temperature would measure 196F. 196F-200F = -4F, which is the correct offset value for my machine. That value would be plugged into the offset menu by the factory, as it is now. Once that's done, the firmware would actually heat the boiler to 204F to produce the desired target temperature of 200F at the group head.

It's not a one-line change in the firmware, but it certainly seems like it would be a simple modification.

Wow, I had no idea the GS-3 worked that way! The Fuji has a single setting to set the offset and it is very easy to do. I don't even know what temp my boiler is actually set to. I just use a scace and play with the offset value (which isn't linear) until the measured temp (scace)matches the input temp (PID) and done! Seems easy and intuitive. Why do it any other way?

mitch236 wrote:Wow, I had no idea the GS-3 worked that way!

Same here. My Vivaldi S1 allowed for easy offset changes as does the Speedster. The Speedster arrives with no offset entered in the PID but a generic -4*F will get you very close. After Scace testing you unlock the PID, enter your correct offset & thereafter just select the brew temp you want letting the software do the computations.

Dick,
I forgot all about this. Please forgive me on the post above. Yes, this should have been done. This was an issue with Gicar since the beginning. It was never solved in 4 years working with them. Not sure why. It did seem soo easy to do a simple subtraction. In the end I think it had to do with the way it was coded for Celsius and the conversion to Fahrenheit. They would have had to rewrite a bunch of their code and I guess did not want to do that.

Anyway now I understand the OP.

BillC

I use the Differential Subtraction method. I know the temperature at the head is 200.8. I want to lower it by 1 degree. I go to the boiler temperature, which is, say 205.3, and lower it by, and here's the trick, 1 degree!

I never have to use the fact that the offset is 4.5 degrees. It's not hard.

Best,

Matt

Billc wrote:This was an issue with Gicar since the beginning. It was never solved in 4 years working with them. Not sure why.

The Strada works the same unintuitive way. I used a Scace II to calibrate, but initially scratched my head wondering why the program temperature didn't change when I changed the offset. In the end, I adopted Matt's strategy and treat the program temperature as "X" and the displayed temperature as the actual temperature.

To make matters even more interesting, I think the offset stays in Celsius even if you change the display temperature to Fahrenheit (admittedly I'm not sure since I treat the offset as "Y"). The offset is -3 on mine, same as it came from the factory.