...split from Compak K10 WBC grinder - user report by moderator...

NoMilkToday wrote:Nicholas: Good to read that your Robur is starting his magic on you. For seasoning a new grinder one truly needs to salvage some beans, two kilos at minimum, I found.
And don't let anyone tell you otherwise: the Vetrano is a fine machine. I understand that you like to modify things. If so, have you considered PID-ing your Vetrano?
Bringing the quality and especially the consistency of espresso on a higher level, in my opinion, an espresso machine needs to have a temperature controlling device, preferably a PID.

Wilco,

PIDing a HX machine doesn't make much sense to me. I know that Ian on the forums did this with impressive results, but I am not sure that I have the interest.

What I have done is simply add a temperature probe to the grouphead using an adapter like the one EricS designed. I get amply accurate brew temperature, and I would say that's one reason why I like Vetrano as much as I do. I have a PID that I am using to only read the temperature at this point. I use it because it's way cheaper and prettier than a Fluke, and it's way faster than anything else I can find (4Hz reading). You can see the adapter setup in this image if you click on a larger version of it... there is a 1/16" tube that comes up out of the top of the grouphead and snakes back into the machine. The PID's power is wired from the power switch on Vetrano, and the power and probe wires are fed out of the machine from the bottom, into the PID, which sits in a project box beneath the machine.

Sorry for dragging things OT. As far as the Robur, I believe it's had about 2 kilos through it as of this weekend, so it should be reasonably seasoned. I don't know if it's that or just me getting used to it, but the pours are more consistent in evenness and flow rate, and the taste is also more consistent... and very nice.

Nicholas,
I'm deeply impressed. You are a man with many talents; from what I can identify: 1) a great photographer, 2) a fine child producer and 3) a creative and neat modder. Your setup looks wonderful!
Do I understand it correctly that you flush water until the desired temp is seen on the PID-display beneath your Vetrano? If so, there are - apparently - more ways to reach the promised land. Learned something new today. Thanks.
Best wishes. Wilco

Wilco, the threads shadowfax linked explain in detail. The short answer is Eric's E61 thermocouple adapter allows you to brew at the desired temperature with minimal fuss. It does waste water, but otherwise it's a "no compromise" solution to on the fly temperature control:



I flushed until a fixed offset of the desired brew temperature; the offset is machine-dependent (it's 4F for La Valentina). The espresso machine's boiler in the video above is running very hot to allow the maximum temperature range with a tradeoff of a long flush and a small loss of precision. Even so, reproducing the desired brew temperature within 0.5F isn't difficult with a little practice.

Dan,
Thank you. Deeply impressed again. I'll dive in the modifications of Eric and Ian for further study. This is excitingly interesting. For people with tight funds and a wish to improve on the consistency of their heat exchanger, buying Erics thermocouple seems to be very good advice ... and relatively easy to implement! Wow. This should be advertised. Ian's PID-implementation looks to be more complex to implement and perhaps a bit more costly (not when you compare it to buying a new machine); the benefit seems to be less water spillage and more ease of use. I like that too, very much even. No good news for retailers, I fear. Life of old machines can be extended significantly.
Thanks again Dan. Wilco

This overly long post is in reference to the comments having to do with temperature management of heat exchanger espresso machines that have worked their way into this thread. If you are only interested in the K10 WBC grinder, per se, you can stop reading this post at the end of this sentence!

As some here know, from my prior interminable posts, I have PID'd two different Cimbali Juniors of different vintages and I use the PIDs to maintain the boiler temperatures at relatively low boiler temps. With this approach, and using a low volume flush (~50ml) I have been able to achieve very reliable and repeatable shot temperature profiles on my machines by varying the boiler temperatures by a few degrees. Realizing that my shot curves are not flat (due to the Cimbali design, which for all I know is intentional, with no evidence existing that "flat" is "better") I've been able to control the obtained shot temperatures well within a range of about 198F to 203F, during the 9 months of the year when the interior temperature of my kitchen is a constant 70 degrees F.

Living in the mountains in an area with few hot days, few people here have air conditioning and my house is no exception. As a result, I rely on "natural air conditioning," e.g. opening the windows overnight to cool the house down (as cold as possible during the hottest days), with the windows then being closed early in the morning. This results in my kitchen having a rather low temperature in the morning (as low as 60F) which then steadily rises perhaps as high as 80 or rarely 84F late in the afternoon. As a result, my espresso machine(s) are exposed to a wide temperature variation during the time I use them, which at its extremes can be up to 24 degrees F. The other 9 months of the year are of no concern as the indoor temperature remains around 70F due to both a furnace and solar radiation.

My own testing has shown that during the periods where the temperatures vary the most, I've been unable to attain consistent shot temperatures with my "usual" approach. In the hope of improving the temperature stability of my machines, I purchased one of Eric's adapters for each machine, and with a little effort installed them and have used a digital thermometer to read the "T-type" probes installed. My hypothesis was that by using Eric's adapter that I could obtain an additional "data point" that would improve my shot temperature stability under arguably very difficult conditions.

During the summer I datalogged many hundreds of Scace shots on the two machines, using both channels of an Omega datalogger. I was monitoring both the real time temperatures on the GH adapter PLUS the Scace device temperatures, and they could be compared at each second of each shot to show how each varied. This included the time before starting to make the shot (e.g. "static GH temperature"), during the flush, and during the following shot. I tried most every possible variation numerous times, such things as flushing to a temperature, waiting for the GH adapter to go back up to a given temperature, standard flush volumes, etc. etc. etc. I still have some data that I have not yet plotted out since it became very tedious and unrewarding after awhile, and I had a lot of other things to do

I tried any number of possible approaches, both with my standard approach as a baseline augmented with knowledge of the baseline (pre-shot) temps on the Grouphead Adapter (hereinafter referred to at the "GH adapter"), to flushing to a specific temperature then starting the simulated shot, to many others. These other approaches included pushing the boiler temperature up much higher than I usually do and flushing to a specific temp, to the best of my ability realizing that I live at almost 6000 feet and flushing to a specific temperature with all the resultant "flashing" of boiling off the grouphead is extremely unpleasant when one lives somewhere that the boiling point is around 201.5F (~94C). I took the boiler temperature as high as 242F, which equals about 0.9 bar on the front panel gauge on my machines. Higher than that just creates so much flashing and superheated water spray that it is intolerable at altitude.

As I stated earlier, I have already found a solution to shot temperature management with my equipment, which works very well under "steady state conditions" of ambient temperature. In a steady state ambient temperature environment, it is possible that the GH adapter would enable me to do a little bit of fine tuning, e.g. to reduce the number of occasional "outlier" shots that don't come out as expected. I have not studied that yet. As to the use of the GH adapter to help with shot temperatures under varying ambient temperatures, I've found that it doesn't really help.

The explanation for this, I think, is pretty obvious. Both the boiler temperature and the GH temperature (not GH adapter temperature) will be primary determining factors in the ultimate shot temperature obtained in an HX machine. The GH adapter measures the temperature deep inside the GH itself, by design near the intersection of the water channel with the stream coming into the GH to make the shot. The GH adapter does not really measure the overall temperature of the large mass of the GH itself where the water shower "rains down" onto the puck in the PF. This is especially true of a situation like the one I was trying to deal with, where the exposed GH itself is exposed to highly variable temperatures in the surrounding air, that the deep channel of the adapter is not seeing (or only "seeing" part of) and hence can't measure. And, once you start running the water in from the HX to make the actual shot, you are most assuredly NOT measuring the temperature of the GH which is exposed to the ambient room temperature which will nonetheless definitely be impacted by the room temperature.

For those few who have read this post down to this point, what is the point that I am trying to make? My point is that IF your inside ambient temperature where the espresso machine is located, varies by more than a few degrees over the course of the year, you may need to check to see if what you measured at one ambient room temperature is what you will get in the cup at varying room temperatures.

My guess is that you will find that you are not getting the temperature stability you (may) have assumed you are getting if external factors such as temperatures change. It is human nature to want to believe that you can calibrate your machine(s) once and then forget about it later, since you already have done the work and of course you did a good job of it I wish that that were the case!

For those of you who "enjoy" constant indoor ambient conditions year round, you will have one less thing to worry about, however there are other things that might get in the way of your (presumed) constant steady-state temperature management (such as, maybe, variations in the temperature of your input water) so things are probably never going to look as "easy" or as "perfect" if you come back at a later point and do a retest. It never ceases to amaze me how the sorts of measurements we take in espresso making, that we assume will remain constant, seldom do.

ken

Wilco--thanks for the kind words; I appreciate them very much. I think Dan covered the general routine for using the GH adaptor with an HX machine, so I think I'll leave it at that so as not to steal his, or any more of your, thunder.

Nevertheless, to wander even more off topic with Ken--
You make a very true observation, that applies to many things in espresso (grinding comes to mind) and life in general. We are always, it seems, generalizing and making more assumptions than we should in order to try to simplify things for ourselves, and it often comes back to bite us in the end.

This, of course, is why we all need to just let it go and enjoy the zen of making a darn good cup of espresso, at 200F or 202F. or 198F. Or, alternately, we could buy saturated grouphead machines like the Dalla Corte [Super] Mini, Synesso, or LM.

In all seriousness, though, I'd be curious to see the kind of data you've accumulated gathered on a) smaller E61 HX machines such as my own and b) double boiler machines with exposed-groupheads like the DoubleDomo, Brewtus, Alex Duetto, etc. My hypothesis is that, over time and medium to large ambient temperature fluctuations, you'd see such double-boiler machines give out fairly similar (to HX) variations in the brew-temp vs. GH Adapter temp vs. boiler temp differential. I would expect even saturated group type machines to give some variations with ambient temperature as well, but I'd imagine it's probably minimal compared to either of the above-mentioned designs.

As I mentioned, though, it's really easy to look at the data you collected and get bogged down in the minutiae of it. On the other hand, I am sure that you don't have any particular disappointment with the quality of espresso you make, or the consistency (forgive me if I am mistaken); I don't have any problems getting good espresso with my Vetrano, although I must admit that repeating flavor profiles exactly (especially day to day) pretty well eludes me. I can repeat the overall profile well, and manipulate it a bit with flush variations, but 'photocopying' espresso shots... I honestly haven't even seen this done at one of the better shops in Houston, much less at my own espresso bar. There are indeed 'layers' of consistency, and I haven't penetrated to the levels of 'verisimilitude.' Still, is that a real problem? I'd be pissed if I were expecting a citrus-melon shot and wound up with a flat, earthy shot despite a good-looking, properly timed shot with the same temperature profile (on the adapter) as a few hours ago. And I'd be at least rather irritated if I ended up with a hazelnut-cocoa shot. But if I'm getting the overall profile that I expect, and the details are jumping around, that seems like more of a 'fun' surprise than a bad one. That's just my opinion, of course; apologies if I misunderstood you, or have utterly failed to make any sense.

Thank you for this most interesting post Ken.
And thank you, Nicholas, for elaborating on Kens post for the benefit of present HX-machine owners.

Food for thought. Based on the information above, I think it's a no-brainer that HX-users with tight funds (and the ambition to serve more consistent espressos) get themselves Erics thermocoupler. Right?

What are the possibilities for present users of PID-controlled machines?
So, I have here a Duetto with a "conventionally" implemented PID-device (measuring from the brew boiler), placed in a room with relatively low temp variation. Suppose I install Erics thermocoupler in the E61 group head (I saw a picture of a version with a temp display on the group head itself; very neat). In use: I imagine using the PID as a baseline and arrive at the desired temp with small additional flushes while monitoring the temp display of the thermocoupler. Question: are there new tast layers to be revealed? If yes, are these clear enough for a "positive business case"?
Feedback based on real experience is greatly appreciated. Wilco

NoMilkToday wrote:Food for thought. Based on the information above, I think it's a no-brainer that HX-users with tight funds (and the ambition to serve more consistent espressos) get themselves Erics thermocoupler. Right?

Personally, I think this conclusion is debatable.

For many machines in most locations, I think that Eric's GH adapter will allow one to achieve repeatable shot temperatures within certain limits. In order to really be sure of one's results, it is necessary to have other sophisticated equipment such as a Scace and a datalogger for confirmation. The last sentence would apply basically to anyone who wants to confirm that they are achieving the temperature stability that they desire, regardless of their temperature control methodology. Therefore, shot temperature consistency does not come cheap regardless of how you try to obtain it.

There are many ways to improve your shot temperature stability. The cheapest approach, requiring no additional expenditure, is Dan's water dance technique of flushing until changes occur in the visible flashing off the group head. In terms of value for money, it is clearly the best

My own approach, of using a lowish boiler temperature and flushing a set amount, works very well in (at least) my machines, and produces results that to my eye are as good as any other, at least based on shot temperature curves I've seen posted on websites such as this one. This approach, at least on my machines, does not work well in the presence of huge ambient temperature fluctuations, although is very reliable in steady state conditions. I had hoped that adding the GH adapter to the mix would further refine the results, but if there is a way to use this information in my situation I have yet to discover it.

On additional advantage of PIDing the boiler of an espresso machine is that it eliminates the pressurestat, which is one of the least reliable components in most espresso machines found in the home. The cheap pressurestats found in semi-commercial machines are junk, pure and simple, and they tend to require replacement about every two years, if not sooner. After a couple of replacements you will approach the cost of a PID installation, and this should be much more robust.

Pressurestats tend to be located in relatively inaccessible places inside espresso machines, and adjusting them requires the removal of side panels. They are not finely adjustable, and are designed as a "set it and forget it" type of component. With a PID installation the controller is going to be on the exterior of the machine and adjusting the boiler temperature is a snap.

Finally, various temperature management approaches are not necessarily exclusive, so you can use more than one technique if the improvements shown are additive.

Whatever you do you should expect to need to confirm and reconfirm your results since they will probably change over time due to factors you can and cannot anticipate.

ken

NoMilkToday wrote:Food for thought. Based on the information above, I think it's a no-brainer that HX-users with tight funds (and the ambition to serve more consistent espressos) get themselves Erics thermocoupler. Right?

That's demonstrably true for E61 groups, but as Ken notes, it may not hold true for other HX group designs. For example, I would not expect it to work well for the Cimbali Junior, which is well known for its heavy group that resists attempts to influence the brew temperature by flushing. Although some have fitted the Elektra A3 with a thermocouple (or two), I believe it's a case of diminishing returns because it's already easy to nail the brew temperature if you have good hearing.

End of flash boiling + six seconds = desired brew temperature

If you need a Scace and a datalogger (or at the very least a reliable digital thermometer) to confirm your "results," then I'd challenge the notion that the purchase of a GH adapter plus this other stuff is a "no-brainer" purchase for an E-61 owner with tight funds. These items added together could easily total $400 or more dollars which is a significant expenditure to most.

I'd question whether the results you get one time with such a setup, even on an E-61, will remain stable over time especially in the face of changing environmental or other conditions, or simply on the basis of "drift." I don't know how often owners of E-61s have repeated their calibrations, between Scace results and GH flush temps shown with Eric's device. My own experience is that variability in measurements with any of this sort of stuff, be it PIDs, the GH adapter, even brew pressure measurements, show up only after repeated measurements over a prolonged period of time (e.g. not just a couple of test rounds done over 2 months).

If owners of E-61s have such data, of repeatedly testing their equipment over long periods of time under varying conditions, showing that the correlations are invariable, I'd love to see it. Simply believing it to be true based on machine designs is a tricky business and I'd suggest a bit more testing before one accepts the notion that these things do not vary or "drift" over time. You may also find that the relationships that hold up on shot series of 1 shot do not hold up when you do sequential shots in that the 2nd or 3rd shot might not have the same relationship between the actual shot temp and what you see on a readout from the GH adapter. The same holds true if you vary your shot intervals, sometimes making a single shot every 30 minutes and other times having much shorter intervals such as 3 or 5 minutes between the shots.

This sort of testing is incredibly tedious and few will want to do it, but you have to do it if you really want to be sure that your approach really works under varying conditions.

Although I have written a lot on this subject of tight shot temperature management, Dan and Jim S. and I are in agreement that this is a subject of overrated importance and that most people cannot taste small (under 3 degrees F) temperature differences in espresso extractions, so to some extent we are debating how many angels can dance on the head of a pin . . . . .

That being said, the $400 or more dollars could probably be better spent by learning to use Dan's "water dance" flush technique (for free) and spending the money on a grinder upgrade.

ken