Silvian lämpötilat - Single Shot

Test I / 17.12.2007

This test was to measure the water temperatures both from the boiler surface and the PF filter with a single shot, around 7 g of coffee. The following picture has also one blank shot of  80 ml from 110 degC and two single shots from 104 degC (the shot is on the left side of the text). The right hand corner shows the layout of Silvia with the thermocouple locations.

The curve also shows how brewing coffee also gives additional activities to the measurements. After the first blank shot, Silvia is also used to wash the equipment. There was enough warm water for both shots, and the heating started some time after the brew.

When we look more in detail what happens during the brew, we get a better understanding of temperature profiles:

The upper picture is the 80 ml blank shot without coffee. The brew starts at 110 degC boiler surface temperature and the PF just below 70 degC. The water release was 13 seconds (+/- 1 s) and the filter temperature increases in five seconds to  91.3 degC and then decreases less than two degrees by the end of the brew. Silvia's heating element starts five seconds after the brew. 

The single shots were started at 104 degC and the coffee temperatures before the brew were 67.7 and 72.1 degC. The beginning of the picture shows, how fast the coffee temperature is increasing as the PF is connected. The scale is deceiving as the increase is 7 - 10 degC per minute.

In the filter, the temperature is increasing again in five seconds to the maximum. This is the same speed as with the blank. The maximum temperatures of the coffee are  89.7 degC and 89.4 degC and only 1.5 degrees lower than with the blank shot. As we note that the water was six degrees higher, it seems that with the coffee cake we get more efficient temperature increase.

As the temperature in the first shot was 65 degC, Silvia cools quicker. However, the temperatures increase in the filter to the maximum in the same time (which also is interesting) and they are the same during the brew.

So, 104 degrees would give for a single shot a brew temperature of 90 - 88 degC. Some comments has said it to be ideal, although also some degrees higher temperatures have been recommend at some sites.

Test II / 26.12.2007

The Christmas went without testing as the theory needed some further study. Also, some simulation programs were discussed. In Ristretto also member K developed scientific theory to temperature profiles.

On the Boxing Day a longer test run was prepared with ten single shots:

The brew flow is calculated by adding to the cup also the water from the trip tray. And finally, the increase in the PF mass after the brew is changed to water volume. As the water content in the cake was only measured in test VIII forward, the average of 10 ml from these tests was added to the previous tests and is included in the figures.

Tests I-VI were with the reversed temperature surfing and VII-X with temperature surfing. For these tests an additional thermocouple was installed under the screen. It was not quite easy as although the thermocouple end could be twisted to be above the screw level, the wire coming through the seal was sometimes hitting the cake. Probably tests II and VII were suffering from this phenomenon as the water flow increased so much.

Also, without intention the first reversed surfing started at 114 degC, 10 degrees higher than intended (noticed only during processing the data). As shown here:

Tests I - IV (reversed temp surfing):
104 degC reversed temp surfing got only one proper test as the first started 10 degrees too high and in the second the cake probably was distorted by the TC wire. The third test gives similar results as above. The filter and group head temperatures are close to each others.

The first 106 degC is also giving similar responses. The lower red curve is from the filter and higher under the screen (click picture to enlarge): 

Tests V and VI:
 Both with  106 degrees. The coffee temperature after reaching the maximum level decrease with less than 1,5 degrees. The profiles look to be between 91 and 93 degrees. The filter starting temperature is almost 75 degrees, although it is not visible in the picture.

Comparing all 106 degrees tests, the profile are looking similar, but the maximum temperature increase in each test with one degC. In test IV the filter temperature is 65 degC, in test V almost 75 degC and in test VI 77.7 degC (click picture to enlarge):

Tests VII - X:
The temp surfing was not as accurate as I would have wished. The temperature increase speed and the data logging cycling made the timing slightly inaccurate. The shown values have an uncertainty of one degC. In test VIII the cake was spoiled by the TC wire as there were almost 2 dl of liquid and the temperatures were very high in the PF. On the other hand, this picture show clearly that the cold water flowing to the boiler lowers the surface temperature to below 75 degrees. The pictures show also the water temperature in the tank and the wait time after the heating has stopped (click picture to enlarge):

Measurement reports
Heating and cooling testing
Blank shot testing
Single shot testing
Temperature measurements - main page