Posts Tagged ‘machine testing’

WBC Testing – Usability and Grinders

Thursday, January 17th, 2008

I am writing sat comfortably in amongst the green coffee down here on Granville Island. I have had enough coffee for a month, and everyone is exhausted and caffeinated.

Despite this, and contrary to what we had planned to do, there has been no official tasting during this WBC evaluation of equipment and personally I think this is a great thing. We had some plans, but very quickly we realised that it would simply be pretty much impossible to run a test that really did justice to the capabilities of the machinery here and to perform evaluations that would withstand both our own, and other people’s, criticism.

Instead we pushed our usability testing a little further and developed the protocols used. If I can get a copy of the sheets we used I will post them. The machine evaluation tests covered things like buttons, ergonomics, the live feedback (gauges, displays etc) machines gave and other stuff (like how easy it is to clean or service a machine). Three groups worked for a set period of time on machines, pulling shot after shot, and then rotating. It has completely changed the way I think about using a machine, and clarified a lot of my thinking on what I want from an espresso machine. Manufacturers were allowed to set up machines as they wished (in terms of preinfusion and pump pressure – more on the latter if I can post about it). We drank a lot of our shots but it wasn’t to rank them – more to see how the machines would respond to us changing the brew recipe or trying to manipulate the machines as we might in competition (flushing routines etc).

Grinders were evaluated in a similar way. Again – we had six grinders to test: A Rossi doserless, a Mahlkoenig K30 and K30 twin, a Compak K10 WBC and the new Compak Doserless “Fresh” and the Nueva Simonelli Mythos.

For grinders with timers the consistency of dose delivery was evaluated, the level of particle aggregate (more on this in the future I hope), how it held a grind and the ergonomics and intuitiveness (is that a word?) of adjustments and controls.

I think the manufacturers are happy and they were around the machines at all times while we worked them, and were helpful in answering any questions. I think, or certainly hope, they are happy with the testing and they will certainly get a lot of feedback from this. If they run with it I am very excited to see what will happen next.

Any testing is obviously open to complaints or the highlighting of problems but everyone involved has worked very hard and I think we are all proud of the testing. The next round in 3 years will really set a standard as we learned a lot about testing machines in the last 5 days.

Don’t expect results any time soon – I don’t know when manufacturers will find out their full results but I do know the winner of the bidding will not be announced til after Copenhagen, which I think is correct.

I will edit my photos on the flight home so I will post when the set is up on flickr in the next few days.

Again I am happy to answer any questions, or to try and clarify how we did things but I am not really comfortable giving my opinion about any of the machines at this stage – certainly until I’ve spoken personally with the manufacturers about the notes on my forms.

WBC Machine Testing – Temperature Testing

Sunday, January 13th, 2008

Vancouver is a lovely city, but it seems it likes to rain. So we’ve all been cooped up inside testing the six machines submitted (I am glad I am not the one paying to run 6 three groups all day!)

People probably want to know how this works and I will try and explain, but I warn you it isn’t massively exciting. As a tester I have no idea how the machines performed. The Scace devices run straight into a datalogger and then into a laptop which is kept out of view. I will never get to see the data, unless a manufacturer chooses to publish. Also a manufacturer will only see their data and not the other machines.

The testing protocol is pretty simple. Each round consists of 14 shots pulled at varying intervals, with the device removed at certain points to simulate grinding, as well as knocking out the puck and flushing. Each machine has four rounds of testing: middle group only, middle group and steaming, 1st and 3rd group only and 1st and 3rd gorup with steaming. With the latter rounds two people worked the machine, even though it was replicating one barista using this. We did this to make sure we could be as consistent as possible in the collection of data across machines. With the steaming we had tested each machine to see the time it took to steam 10oz of milk up to 60C, and then that time was used during the testing.

It hasn’t been the most fascinating testing to watch – every manufacturer was present with a technician during the testing and I think the constant counting down each action, second by second, was close to driving everyone a little crazy!

I think the testing has been very fair, I think the protocols are very good (though I don’t think anyone is claiming that it is beyond improvement). If anyone has any questions I will try and answer them though if you have issues or complaints I am probably the wrong person to speak to. Everyone seems very happy with the fairness of it, which is good.

Next up is grinders which will take a while, and I will write a bit more about that after we have done it.