Video: Understanding The Siphon

Hello from James's team! We've uploaded the newest YouTube video here for you, ad-free. We dive deep into some data collection and experiments in an effort to understand the most inscrutable of coffee brewers: the vacuum pot or coffee siphon. We found some super interesting stuff. This is the second in the siphon series, and the next stop will be our ultimate technique. (Though, as a warning - that might take a while. This video might have raised more questions than it answered...)
Here's some behind the scenes perspective from Sarka, who worked on the testing for this video:

Hello, fellow weird coffee people!

It’s been a while since I provided some commentary on our testing, and this syphon video seems to be the perfect one to share some thoughts about.

Let me start by admitting that, prior to this video series, I had never brewed with the syphon before. It is not the most convenient apparatus to brew with, no matter where you are. Busy cafes, especially, might find employing the syphon a bit tricky. However, despite its many parts and the involved process, you can still find them in cafes around the world.

I do have a vague memory of when I came across it for the first time. I was freshly 18, working my very first coffee job as a part-time barista in a speciality coffee shop. There was an old, dust-covered unit on the very top shelf behind the bar. Without the filter, top cover, and burner, it served as decoration rather than a functional piece of equipment. It looked like something used for lab experiments, not coffee preparation. I thought it was the coolest thing one could ever brew coffee in. Sadly, I never owned one. Shortly after I started to work in that cafe, I was riding the V60 express, and the syphon solidified itself in my mind as more of a prop; an interesting, seemingly fun and complex “coffee thing”. When we decided on doing the syphon series this year, it was exciting to finally try and brew with this legend of a brewer.

Before any actual testing could even begin, a couple of things needed to be figured out. James’ testing protocol involved temperature control and information on the pressure in the lower chamber during each brew. Hence, we needed a lower chamber that would allow us to collect data on both. A round-bottom borosilicate glass flask (repeat it fast three times) with 3 necks did the trick, with some adjustments needing to be done. So I bought some diamond-covered disks for our Dremel, read one Reddit post on cutting tube-shaped glass, and went on to practice on a spare round-bottom flask we had purchased. I like talking about this little detail of this project, as I’d have never thought I’d be cutting glass at work, and that it would be such a fun thing to do! You can enjoy a short snippet of the glass cutting and shaping set up in the videos attached to this post. If you feel like questioning my safety equipment, just know that 2 syphons were harmed during this process. Not to mention, I am probably the most frequent user of our first aid kit.

Once the round-bottom flask was finished, and a temperature probe along with Smart Espresso Profiler were fitted, I could start testing.

Test 1 of the protocol was focused on varying temperatures of the bottom chamber when sealing and the effect it has on the top chamber. If we seal at lower temperatures, will we see lower temperatures in the top chamber? If yes, is there a consistent temperature drop, and can it be predicted? An interesting observation when working with the Frankensyphon was that water in the bottom chamber doesn’t start rising to the top chamber until it hits around 95°C. That seems to be the point where there is enough pressure built for the water to be pushed up. No matter if the chamber is sealed at 80°C or 93°C, the top chamber temperature corresponds to the temperature we seal at, not to the temperature the water in the lower chamber had to climb to after being sealed. This doesn’t replicate the same way on a classic syphon. The water always started moving immediately after sealing. The difference was that when sealed at a lower temperature, it was moving slightly slower than when sealed at a higher temperature.

Test 2 was about whether top chamber temperatures can be controlled. Using the watt meter, with the halogen lamp at different power percentages, I tried to see what the correlation is between wattage drawn and the temperature profile in the upper chamber. Test 3 repeated Test 2, this time with coffee added to it. We observed a larger temperature drop over 2 minutes when the heat source was set to a lower power setting, with a temperature increase when the halogen lamp was around 70% of its power and higher. And finally, our Test 4 focused on pressure inside the lower chamber during brewing. All of these points are addressed in the video, so I will let you enjoy James’ commentary without repeating it here.

The vast majority of what we have found has been shared by James in the video. But during filming, we found out that more questions had emerged. One example: To speed up the preparation of individual brews, James would use a cold cloth to cool the bottom chamber down faster. Cooling the bottom chamber results in the coffee being pulled back down faster, but does it affect the negative pressure peak? I went and measured a few brews to see how rapidly cooling the bottom chamber affects the lower negative pressure peak due to the liquid being forced down faster.

The bottom chamber was filled with 250ml of water (off boil from a kettle) and sealed when the temperature in the chamber reached 95°C. Once the water climbed into the top chamber, the halogen lamp was taken down from 100% of its drawn power to 70% for the remainder of the brew, and 16g of coffee was added to the chamber. I did a 1-minute brew with a stir at 30 seconds, and a gentle clockwise and anticlockwise stir at the end. The results of negative pressure readings are below.

The control here was the negative pressure produced in the lower chamber when the halogen lamp was turned off, and the cooling process happened naturally. I then proceeded to fill a bucket of water and put a sous vide in to keep the temperature stable whilst I brewed coffee. This time, after the halogen lamp was turned off, I took the syphon and submerged the bottom chamber under water. This was done 3 times at 60°C, 50°C, and 40°C. There is a clear dip in negative pressure corresponding with the water temperature in the bucket. An interesting observation at the 50°C and 40°C mark when cooling the chamber was the bubbling sound when the air is forced back into the lower chamber after the last bit of coffee has been pulled down. This is something that happens when brewing with the classic syphon all the time, but wasn’t happening with the Frankensyphon during testing. It makes me think that it is a more accurate representation of negative pressure occurring in the classic syphon rather than the readings we were able to collect during testing.

I hope that you find this bit of information interesting. I definitely did! And I will be spending a bit more time on this to see if the extraction is affected by the higher negative pressure pull.

Please, if you have any ideas on what else you’d like to see regarding the syphon, let us know. In the meantime, we’ll be working on developing the ultimate syphon technique that produces delicious coffee with every brew!

Until next time,

Sarka