OK, I've got my hottop set up and connected to my laptop via an omega HH506RA. I'm monitoring and recording bean and environment temps. I can find the output file, and after the data is done being logged, import it into excel and graph the roast. But I can't access the file during the roast, because it is being written. Is there a way to grab the data and put it straight into an excel graph as it is being written? I've been trying to figure out microsoft sql servers and oddb sources and all this computer stuff and i have no idea. anyone doing it successfully? Also, does anyone have a nice spreadsheet they use for the purpose? One that will accept an omega HH605ra output would be a bonus. I'm not bad with excel, just lazy right now.

I can't answer your question directly, hopefully someone can, but I can suggest an alternative:

If you happen to have MATLAB on your computer you can use it to read the data from the thermometer directly, bypassing the Omega software altogether. Here's some sample MATLAB code that reads the temperatures and plots them on the fly, as it happens:

s = serial('COM1','Terminator',char(13));
fopen(s)

t0 = tic;
for i=1:56,
    data = fscanf(s);
    T(i,:) = [sscanf(data(12:15),'%d'), sscanf(data( 8:11),'%d')]/10;
    t(i) = toc(t0); plot(t-t(1),T,'o-'); xlabel('TIME (seconds)');
    ylabel('TEMPERATURE (F)'); legend('T_1','T_2'); drawnow;
end;
fclose(s)

Here's the resulting output:



I am sure this could be done with other software packages too, MATLAB is just the one I am used to.

John

Some multimeters seem to be designed with the limitation of not being able to directly access the data table on the fly (or more accurately not being able to access the output of the probe status through USB in real-time as it's designed to first dump to a table and then retrieve it later as you've found out). These are usually the cheaper USB meters that generally don't say one way or another about real-time data reporting. I think the problem is a limitation in their programmed firmware, which could potentially be rewritten and flashed, but pretty unlikely unless someone knows the hardware very well.

Hopefully JB130's suggestion of being able to hook it directly to the computer should work. I assume that would be instead of the multimeter. Does matlab have capability for using both thermocouple and thermsistor signals directly? Or just one type? If so how is it wired up to a serial port (or maybe more ideally USB) connector?

Another alternative is looking for a different meter (usually more expensive), but I hear the folks over at coffeesnobs have a decent priced one that they've written their own software to use the real-time data to log onto a laptop while roasting.

Thanks for the info!

Droshi wrote:Hopefully JB130's suggestion of being able to hook it directly to the computer should work. I assume that would be instead of the multimeter. Does matlab have capability for using both thermocouple and thermsistor signals directly? Or just one type? If so how is it wired up to a serial port (or maybe more ideally USB) connector?

erics wrote: To acquire data directly from the probe is a great suggestion but that requires an interface card or device for your laptop - right?

I can see now that my original post wasn't clear: I still need the meter itself, it's the meter's software that can be bypassed. The Omega thermometer spits out temperatures in a simple ASCII string at regular intervals from its serial port, so if you have any way of reading directly from a serial port -- MATLAB is just one example -- you can read from the meter directly. In one of my google searches I even remember stumbling across some homemade C or C++ code for this purpose.

John

thanks for the help. I'm checking out the linked spreadsheets now. I've developed my own over the past couple of weeks. Still no realtime excel graphing, although i also found the graphs output by the HH506 software itself. I think i need to get into C++ or at least visual basic to do the work, and i haven't looked at that stuff since high school. It's a shame, because I'm pretty sure you could grab some neat realtime indicators that would help predict where the roast is going pretty well. I'm specifically thinking of derivatives and integrals of the time/temp relationship. As it is now, i do a quick derivative in my head (basic slope calculations). But second and third derivatives would help you catch and predict where the roast is heading much more quickly. There seems to be about a 45 second lag between heater changes and my ability to notice ET/BT changes, but that probably could be greatly reduced with a nice spreadsheet and the appropriate equations. Integrals (at least the first integral) could be a neat way to approximate total energy put into the system too. (adjusts geek glasses, leaves for another roast).

That would be a pretty cool idea. On some aircraft nav displays they have 'noodles' which are this kind of thing. They show the instantaneous rate of turn projected into the future at set intervals from the aircrafts present position - 15, 30 and 45 secs I think. As you hinted at, you could plot the slope prediction in real time and see where it falls against the slope you are aiming for, and adjust/anticipate very quickly.

I do that (plot the first and second derivative) with some of my excel graphs for roasting @ http://github.com/blowery/roasts/tree/master It even updates on the graph as you go. I hand enter times every 15 or 30 seconds, using a little timer that beeps once on the 15s. Gives me a rough idea where roasts are heading. Feel free to reuse these as a template...

Hi
I have the coffeesnobs USB meter. You just connect the meter to a PC, and then run the software that they provide. While running, it gives you a graph of the roast, and tells you the current rate of change in temp, in degrees/minute. When the roast is complete, it will save the data to an excel file. You can copy and paste the data into a another excel file, if you want a post roast chart of the data.

I think you could go there, and ask for the source, and see if you can modify that program to read data from your omega. The existing sw was written around this meter, but it's something to consider.

You can read about the meter and sw here
http://coffeesnobs.com.au/YaBB.pl?num=1236075603

Randy