One for @"DavecUK"#1 I suspect. Thanking all contributors in advance. I'm establishing a "grinder lab" to compare burr sets side by side. An important part I have two Puristka's equipped with Geisinger temperate probes in the group heads. I'll upload a video in a new post below, but in the meantime my challenge is to how to complete simulatenous pours one the two macines and have the temperature to within one degree through the group head on each machine. This is more challenging with a machine with a small boiler because the window of stable temperature is limied. You'll see in the video below and the temperature through the group head, as measured by the Geisinger probes, initially riases and then falls dramatically, presumably as the heated water in the boiler is replaced with water from the external resevoir. My question is: whats the best method for having both machines pour the espresso as close as possible to thesame temperature? Note: I only need to do one pour per day simultaneoulsy on both bachines, not back to back pours.

Here's the video, in case that helps. https://youtu.be/hseUXhnTNfM

Your video is a "frozen" photo 😊…edit…it was only a temporary issue with the video loading or else, anyway now is ok. OPV at same setting, push the pump button without to move the lever until the OPV will discharge, so you will fill the boilers at almost the same level, of course all the PID parameters must be the same, wait at least 1 hour before the first shot so you are sure that the whole system has got a stabilized temp. I did follow such simple method with my Carola, and for me was ok to reach a quite stable condition for subsequent test, but i never compared 2 of it, anyway the Carola has a different thermosyphon path, so maybe it could differ in the behaviors.

What's it like with coffee in the pf? Any slower to swing?

> @"giampiero"#p42678 push the pump button without to move the lever until the OPV will discharge, so you will fill the boilers at almost the same level The brew boiler is always full.

@"giampiero"#p42678 oh that sounds promising. Nothing like the voice of experience Thank you! PS I accidentally uploaded a photo instead of the video, I edited it quick smart but you may have been the victim of my initial e-tardness.

@"Cuprajake"#p42680 temperature swing? good point. very good point. will check it tomorrow and reshoot the video. thanks for the idea. It could slow the fluctuation down.

Yeah possible., How much did you flush through in the vid? I know it's not an hx so not sure if a flush applies, on my old hx I would flush till the temp peaked. Then pull a shot,

But really you shouldn't have too. @"Beverast"#66 has one of these I'm sure And @"DavecUK"#1 maybe a better pid setting?

Would there be a difference when water is just free to run through, vs pressurised ? In the former, the heat is being taken straight out, whereas in the latter (usual brew) there's a much slower actual flow and added backpressure that could mean the heat loss is much less.

@"mobius"#p42694 Yeah I would make that assumption as well. I have not done tests with external thermometers on my Puristika but under pressure the shots always come out nice and hot…usually have to let the brew cool for around minute or so before I can drink it. Would be good to see what the readings are with the pressure.

ECM Puristika temperature challenge

MediumRoastSteam

giampiero yes and not, especially with a machine starting from a cold condition

Good point. 👍

tompoland

Cuprajake

Indeed, it’s not really about each machine hitting a specific temperature, it’s about them hitting the same temperature. Frankly, even if they both read 93 at the boiler and are pouting at 90, that’s fine because it’s not a cupping, it’s arout comparing two different burr sets

Therefore, so long as all variables are within a fraction of each other, it’s mission accomplised.

tompoland

giampiero check what temp you get at the group head thermometer during the shots?…or i missed something

that would be too simple. 😀

seriously though, the reason is so that the temperature readings are visible, in real time as they pour, without having to manually measure them.

My OCD method is:

PID controlling the boiler temperature using the same make and model of high quality espresso machine to maximize the likelihood of consistency between the two machines
Measure the temperature of the water passing through the group head
Measure the in-cup temperature once the side by side pour has finished.

Once I’ve confirmed that the temperature is consistently similar across both the cups I can probably dispense with the last step.

Thanks for your help!

tompoland

DavecUK thanks Dave.

Yes calibration has been checked and I can confirm that both probes read teperature indentical to each other. And that’s the main thing I’m after.

Acheiving a similar temperature in the cup should be fine because the cups are identical, everything is in the one room, neither machine is next to a sunny window or anything like that. Beans will be the same and so on.

Don’t think I want to start using a Scace and I suspect that for the purposes of the grinder/burr comparision lab thingy, it’s not as important as it would be for say checking the accuracy or pressure/temperature on any given espresso machine.

As always, your input is noted and appreciated.

I’m off to dial in 2x Varia VS3s and video a simulataneous pour from the two (basket-filled this time) Puristakas. Will post the video here again.

Thank you to EVERYONE for your input. Your support in figuring this out is exceedingly helpful.

tompoland

Back again. After three hours of dialling in grinders, and figuring out that my film director skills suck, I have a video.

TLDR:
00 seconds temperature difference is 0.6c
13 seconds temperature difference is 1.7c
22 seconds temperature difference is 2.3c
31 seconds temperature difference is 2.7c
Average difference in temperature is 1.8c

My idea is to lower the boiler/PID temperature in the “hotter” machine by 2 degrees and repeat the exercise tomorrow.

Anyone got any other ideas? Thanks again!

DavecUK

tompoland tom before changing anything on the machine, try again….Then give the machines some standing time and swap the temp sensors…..see if it follows the sensor!

mobius

So does an initial 0.6 temp diff (and perhaps different internal tolerances affecting flow rate etc.) account for the huge difference also (~10s) in pour time, or are you now in the realms of differences in tamp and other variables also…

Have you tried swapping the machines around in case location is affecting things ?

Either way, you’re brave, crazy or probably both, trying this 😁

The most important question though - how did it taste.

tompoland

mobius probably more foolish than brave. But definitely a bit weird.

Good idea re swapping the machines around. First though I’ll see how I go tomorrow with dropping the PID by 2 degrees on the hotter machine. Now that I think about it, if it is machine location then I might confirm that is the issue if I swap them but I’ll still be left with the differential.

The pour times are due to using different burr sets/grinders for each machine. When I have another go tomorrow, I’ll use just the one grinder which theoretically will sync the pour duration. Not that is important I suppos it just annoys me 😒

Cuprajake

force tamper would sort that out

tompoland

Cuprajake dont think that there is an issue with the tamper or puck prep but appreciate the idea.

Two different burr sets so that will often result in two different pour times.

tompoland

DavecUK ooh I DO like that idea Dave. Maybe three goes without changing anything?

And yes, of course! Swap the sensors!

Danke danke danke!

tompoland

@DavecUK thinking about swapping, I did note this morning when I turned the machines on, that the readings on the Geisinger probes were identical at 21.9c.

Given that, is a swap still worth trying?

Not that it’s much effort. I’m just always slightly terrified of stripping threads. I have form.

DavecUK

tompoland Given that, is a swap still worth trying?

It’s something I would do before playing with machine settings

giampiero

I’ll try again but swapping the machines in respect of the burr set, I once verified that a faster flow it can contribute to an higher temp at the grouphead.

tompoland

Completed another pour tonight (now-ish). As per @DavecUK ’s suggestion I changed nothing other than to swap the Geisinger group head temperature from one machine to the other.

(Please note that a different grinder was used for each pour, hence the difference in pour time.)

A much closer range of temperatures, with the maximum difference being 0.6 degrees. I can live with.

The big question is “why was the maximum range 2.7 degrees yesterday and only 0.6 degrees today”. If the difference was in the machines, how has is rectified itself overnight simply by swapping the Geisinger temperature probes? If the difference is in the temperature probes, ditto.

I’ll run the test again tomorrow and see if the results are still stable.

Here’s the numbers (video underneath)

DavecUK

tompoland The big question is “why was the maximum range 2.7 degrees yesterday and only 0.6 degrees today”. If the difference was in the machines, how has is rectified itself overnight simply by swapping the Geisinger temperature probes? If the difference is in the temperature probes, ditto.

Did you use the same grinders and coffee in the same machine number and get the same pour times/weights as before?

tompoland

DavecUK pretty much:

same beans, same grinders. I has cleaned the grinders (mistake?) earlier in the day and had to dial them in again. Not sure on the result out but it would be pretty close.

I will repeat tomorrow.

giampiero

tompoland the last video you upload, it seems the same video of yesterday, ( i did stop both videos at 10/20/30 seconds and look like identical) that’s why the “numbers” you posted does not match😉

tompoland

giampiero yep, well spotted that man. someone get me a director and producer please. Give me ten.

The numbers I posted above for the Day Two pour were correct, the video was the old one. it was a file name error. Give me ten please.

tompoland

Correct video has now been uploaded. Confirming also that the numbers originally posted are correct.

Tomorrow I’ll use just the one grinder which will eliminate a variable.

« Previous Page Next Page »