EM6910 v Silvia v the world!

[JavaB]

Re: EM6910 v the world!

When I did the same test on the La Cimbali and the Solis SL90.... (a HX and a single boiler respectfully) I got 93C on the La Cimbali for probably 120+ ml then dropped a fraction of a degree from there - and on the Solis 93C for about the first 40ml then it started to fall quite quickly (presumably cold water mixing in the 300ml boiler - and the heater couldnt keep up).

For my measurements I used the flexible bead probe so there was a very small mass to get up to temperature..... and the cup was cut off so the bottom was roughly where the bottom of the PF would be.

I also tried the probe right at the point where water exited the shower screen (on both machines) - and found the previous temperatures to be the same - so I dont think the stream of water (high specific heat) falling through the air (low specific heat) cools significantly.....

The main problem I see is rate of flow (just like the HX).... if the flow rate is too high, less heat is transferred to the water, resulting in a lower temperature (which will be measured)

[Wushoes]

Re: EM6910 v the world!

I have the flexible/wire type probes too, perhaps I should have used these probes instead?

In either case it doesnt matter as it seems the temperature readings are unreliable.

Main conclusions:

1) Upping the brew temp results in a....drum roll....higher temp
2) Shot stability is quite good.

Anybody else want to try this on their Sunbeam to compare to my results?

[JavaB]

Re: EM6910 v the world!

Yep,

Those flexible probes respond far faster. And dont absorb anywhere near the same amount of heat.

Just a thought.... it would be really useful if you could measure the temp of the actual thermoblock... after all that is the maximum temperature the water could get to - and in fact should be pretty close to the actual temperature of the water - but that probably requires some major disassembly .

[luca]

Re: EM6910 v the world!

At the specified temperatures for brewing the combined error is +/- 0.75%

Nice! But that would be like 0.6C, which intuitively sounds like way too little, based on Bichts thread linked to above. He gets 0.7C repeatability with a $1500USD meter and an expensive specialised probe, so I think that thats a bit ambitious! His formula seems to be to add the error to the error for the probe itself, which should be something like 2.3C for a K-type probe. So Id guess that were looking at +/-3C or thereabouts. Which is pretty good, but obviously can have a huge impact if were talking espresso temp.

*Flow rate (ie. if it was too fast, youd expect the temperature to be too cold. Also, apart from the machine being a thermoblock, I guess that water pooling in the cup would probably have some sort of effect). Did you check to see what the flow rate was?

Around 1 to 1.2mL/sec, about the same as the flow rate of espresso pouring.

Hmm ... did you end up with like 75mL total or so? Thats about the amount that goes into making a double espresso, including absorption by the puck.

I would think this is where the greatest losses occur to the environment. This is by no means an ideal experiment as:

1) Brewing pressure is not simulated.
2) Water has to fall a significant distance from the group
3) There is no material that has similar temperature absorbing properties as coffee.

Yep, well we could improve it a little bit by cutting the cup to make it fit snugger against the group. This would be interesting to do in itself, if you can be bothered, because it would give us an idea of how much heat the water is losing to the air. I dont know what difference pressure really makes, but I would have thought that if the flow rate is right, the water will absorb heat from the thermoblock at the intended rate. If you want material with similar thermal properties to coffee, the thermofilter just uses PVC, I think!

Cheers,

Luca

[JavaB]

Re: EM6910 v the world!

Nice! But that would be like 0.6C, which intuitively sounds like way too little, based on Bichts thread linked to above. He gets 0.7C repeatability with a $1500USD meter and an expensive specialised probe, so I think that thats a bit ambitious! His formula seems to be to add the error to the error for the probe itself, which should be something like 2.3C for a K-type probe. So Id guess that were looking at +/-3C or thereabouts. Which is pretty good, but obviously can have a huge impact if were talking espresso temp.

Luca,

You can quickly test using boiling water as your reference- and so long as you are at sea level - or pretty close to it - it will boil at 100C

I have two DMMs with temp probes (one of which is the one Andy organised for us) and both read within 0.2 degrees of 100C (both low by the way).... Using the supplied K type thermocouples.

So pouring boiling water into the test cup would plot the response time for the probe and its accuracy actually in the foam cup.

[Wushoes]

Re: EM6910 v the world!

Yeah. I guess we should be talking about 3 deg or thereabouts.

My guess would have been more than 75mL for sure. I simply pressed the double cup button on the machine to give me repeatable volume.

One of the first things I did was check to see if the cup fit under the group. It didnt, so I cut it down to size. It was a nice and snug fit, propped up by a duralex glass.

[Sparky]

Re: EM6910 v the world!

Getting the flow rate correct on a thermoblock is crucial. IMO Alan Frews test is doomed to low readings and represents pretty poor experimental science. It was really like comparing apples to oranges when he compared his results to a Silvia, as boilers respond to high flow rates very differently.

Short of making up your own thermofilter device (pretty easy with a lathe and drill press) the best way of getting the brew temperature is to snake a thin TC wire over the lip of the basket. I have found minimal leakage when using this method and the TC is seeing the water temperature right on top of the puck, and the flow rate is correct for the measurement. So if anyone has a thin bead type TC with a teflon jacket (K-type available from Dick Smith), then youre set to go. No need for a thermofilter.

Cheers,

Mark.

[chris002200]

Re: EM6910 v the world!

Luca,

It would be good to see what yourthoughts on teh 6910 are after spending a bit of time on it pulling shots and texturing milk.

I know of a few people in Perth who work on some top end commercial machines all day and are still impressed on the 6910s capabilities.

Make sure you give us your feedback should you get a chance to play.

[chris002200]

Re: EM6910 v the world!

Is anyone using different baskets to what was supplied with the 6910?

I have heard of people uing the old 6900 "quad" basket.

Any advice from people on any other baskets worth trying?

[marcstolk]

Re: EM6910 v the world!

The Quad baskets allow for more errors in dosing compared to the standard doubles... otherwise, youre just wasting coffee imho of course ;-)

[chris002200]

Re: EM6910 v the world!

So what baskets are you using marcstolk

[marcstolk]

Re: EM6910 v the world!

Smaller ones that come with the EM6910

But even they are a tad larger than "Italian Standard"

[chris002200]

Re: EM6910 v the world!

Yeah. I have resorted to using the "double" that comes with the 6910.

Thinking of trying some different options from different manufactures

[marcstolk]

Re: EM6910 v the world!

Yeh, me too

[chris002200]

Re: EM6910 v the world!

Im still to decide on what baskets I should try though. I guess it will be a bit of trial and error.

Any suggestions on brands and sizing