Cimbali Jr = sour everytime, why? - Page 2

[torretta]

Malachi, yes, and that's why I almost always go with the espresso blends and roast according to SM guidelines. Thanks for your fine advice.

SM guidelines often produce roasts that is on the lighter side which aim to preserve the characteristics of the bean. Try roasting a batch of whatever deep into the second crack. Your roast color will be much darker than previous roasts with possibly some spots of oil.

[SamuraiE]

Big thanks to everyone who has given advice on this topic. Another question: contacted the service center for La Cimbali in Japan and they will be more than happy to come look at my machine for $250 (that's just to come to my house, no promise of fixing the sour problem).

So I would like to ask if you think it may help if I buy a voltage booster which will ramp up the power from my outlet from the standard 100v to 115v. Checking my line with nothing else running my home's current is a stable 100v (that is what it should run at as well as my machine which was built to run at 100v).

However as soon as I plugged in another electronic device the line went to 95v. Do you think this may be the problem with my cold, sour shots? No one around here sells a variac but rather a voltage booster which is supposed to keep the current running at 115v, give or take a few volts. If I run the machine at say 110v is that going to zap out my motor or anything else? Sometimes the first shot is fine but the second is cold and sour, could this voltage issue have something to do with my recovery time?

Trying to stay away from the $250 home visit, any thoughts would be helpful.

Key

[HB]

Do you think this may be the problem with my cold, sour shots? ... Sometimes the first shot is fine but the second is cold and sour, could this voltage issue have something to do with my recovery time?

No, I don't see how it would matter.

The boiler temperature is controlled by a pressurestat and heating element; the heating element's efficacy is impacted by lower voltage, but only to the extent that the time between the bottom and top of the deadband may be elongated. For example, say the heating element clicks on at 0.75 bar and turns off at 0.9 bar, taking 15 seconds versus 5 seconds to recover. I wouldn't expect that to have any measurable effect, let alone the marked second-shot sourness you describe.

I hesitate to speculate on the cause because your earlier temperature measurements weren't conclusive, both in the data points themselves and the means of collection (i.e., a "Styrofoam cup" test is at best accurate to a few degrees). If there is a mechanical cause, I'm still betting on the heat exchanger injector since I have seen the behavior you describe when I worked on the German-market Andreja Premium (i.e., for a Cimbali Junior, I would expect a shorter heat exchanger injector = naturally cooler extraction temperature). Whether it is worth an expensive home visit to pursue this guess isn't an easy decision. If I had a ream of thermofilter-type quality data supporting your observations, I would have busted out the wrenches long ago.

If do-it-yourself repairs sound like an intimidating proposition and you're convinced you've eliminated the "operator error" possibility, I agree that a housecall would be the best bet.

[rfc]

I've read and re-read this thread, and must weigh in on the side of SamuraiE's suspicions about voltage. Despite the insistence by some that "voltage doesn't matter", I would suggest that all the evidence (especially the fact that when the machine is maxed out he gets a good shot) points to the possibility that it does matter.

For whatever small investment one of these would take:

http://www.toddsystems.com/bt.html

I say: Try it.

Suppose the folks who sold him the machine did NOTHING to it except put a new label on it? If voltage doesn't matter, what would they change anyway? And what would it hurt to give the thing a solid 115 volts.

Another concern is not the voltage, but frequency. If it is 50 hertz, which it is in some parts of Japan, that may affect the operation of some of the components. I am not familiar with this type, but certainly the pump would be my first suspicion.

My vote is to throw a boost transformer at the thing and see what happens.

[HB]

Despite the insistence by some that "voltage doesn't matter", I would suggest that all the evidence (especially the fact that when the machine is maxed out he gets a good shot) points to the possibility that it does matter.

Hmm-m, let's approach the question a different way, ignoring voltage for a moment: What is the behavior of the boiler pressure gauge readings compared to a "known correct" Cimbali Junior? My limited understanding of electricity is that reducing the voltage available to the heat element means less power, which would be effectively equivalent to replacing it with a lower-resistance element. Whether a machine is 1200W or 1400W shouldn't matter. The reduced frequency will indeed slow down the motor, but I understood he measured the brew pressure, so I eliminated it as a factor.

To be honest, what makes this diagnosis difficult is the lack of reliable information.

[SamuraiE]

Dan, I agree with you on the lack of reliable information, so...please tell me what info a troubleshooter (such as yourself) needs (remember that I am not an engineer and have little mechanical knowledge, but don't mind looking inside and taking stuff apart). What readouts would be helpful? If temp outputs are necessary what would be the best way to get them?

I am in the act of buying a variac (not an easy purchase here in rural Japan) and if it does nothing for my machine I can at least play around with it when roasting on my Hottop.

Do I understand you correctly when you talk of the injector as the tube that runs into the boiler? Would you suggest pulling apart the brass group in order to get a look at the injector running into the boiler to see if it is indeed too short?

I'm game to roll up my sleeves and get to it.......just tell me what to 'get to'.

Thanks, Key

[HB]

What readouts would be helpful? If temp outputs are necessary what would be the best way to get them?

Very good question. As I said earlier, now is a good time to call in a local technician, as we're guessing at the cause, and you don't have the tools to eliminate the guesswork. It's an expensive visit, but it would answer the mystery of the second sour shot.

That said, if it were my machine to play with, I would use a thermofilter or take a few dozen in-basket thermocouple measurements. Those are reliable and accurate enough that we could say with confidence that the problem is mechanical, not operational. Since you don't have those lying around, I'm racking my brain trying to suggest a low-tech temperature measurement test that would prove or disprove that the profile is consistently declining after the first shot.

How about this...

You could try to simulate flat-out shot pulling: Get your carton-and-thermometer setup you described earlier. Program the volumetric dose to 90ml. Without flushing at all, draw and measure water temperature at 30 second intervals. At least 20 times, preferrably 30 or more. The first four or five measurements should be high (e.g., 97C) and then I hope they'll plane out around 92-93C, they may even rise a bit. If on the other hand the temperatures trend lower and lower, then I'll start to believe the cause is mechanical.

I am in the act of buying a variac (not an easy purchase here in rural Japan) and if it does nothing for my machine I can at least play around with it when roasting on my Hottop.

I'll be eating crow if that fixes the problem, but that's fine with me.

Do I understand you correctly when you talk of the injector as the tube that runs into the boiler? Would you suggest pulling apart the brass group in order to get a look at the injector running into the boiler to see if it is indeed too short?

Yes, the heat exchanger injector ("plastic pipe") is pictured here along with step-by-step pictures of how to reach it. My wild guess is that the injector has come loose, although honestly I would fall off my chair in shock if that turned out to be true. Even if you took it apart and measured, I don't know enough about the varied lengths to confirm it's any different than those in the US. So it is very likely a wild goose chase. Heck, I would have guessed scale in the heat exchanger had you not said it was brand new...

PS: Suggestions from the engineers among the membership are welcome. Hello Lino? Greg? Sean? AndyS? Barry? Jump in anytime!

[Ken Fox]

Do you think boiler setting is going to make any difference?

Junior is very thermally efficient, so the best boiler setting is lower than you might expect. I tried the range from 0.7 to 1.4 bar; 0.9 bar was the setting when the machine arrived and it worked best for me. FWIW, Ken lives at high altitude and his setting is even lower (I believe 0.7 bar).

Actually, my boiler temp is no longer controlled by the pstat; about 1.5 years ago Jim Schulman helped me to PID it. I have my boiler temp set at 242F as I type this, which corresponds to 0.95bar on the front panel gauge. I would argue that this is more or less the same average idle temperature as one would find with a stock, pstat controlled, machine set to an upper limit of about 1.02 bar with a lower limit around 0.87 bar or thereabouts, assuming a deadband of 0.15bar or so. I have used the machine previously with a PID setting of 236 degrees F. The only real difference was noticeably less steaming power, important since I make some cappas in the morning.

It has been explained to me before the impact of altitude on FPG pstat readings, however I have temporarily forgotten it. In any event, the thermocouple produced readings I get on my boiler temperature should be nearly absolutely accurate as they do not depend in any way on altitude. Suffice it to say that for any given temperature in the boiler, one is going to get more flashing at higher altitude than at sea level given the change of the boiling point of water with elevation increase. If I lived at sea level I could increase my boiler temperature and get less flashing for a given temperature.

I think that the proper setting of a pstat or thermocouple/PID will vary by individual, depending on their preferred coffee blends, roasting level, interest in frothing milk vs. simply making espressos, and of course, personal taste. I would test different flushing approaches and see what worked best for me with a given machine in a given location. I flush around 120-135ml when the machine has been idle for any period of time, but that is just me. I have that 130ml-ish flush programmed into one of the buttons so it is automatic for me in my espresso preparation.

ken

[SamuraiE]

Well, I doubt Jim Schulman will be coming by to PID my machine, so I continue on this journey, with the help of all of you.

Testing: Did the best I could with my lame testing setup, $20 dig thermometer (made in China) and my 1 liter milk box stuffed into a coffee mug.

Pours were about 8 seconds which gave me 165 ml or 5.5 oz of water....didn't use a pf, just poured straight from the group right into the carton. Wanted to draw less water but didn't realize until too late that I was getting such a high volume. Pulled for about 8 seconds, waited 30 seconds then pulled again, and again, etc. (as per Dan's orders).

Sat. night (late...1 am) machine boiler set at 1.6 bar and group brew pressure at 1.1 bar:

1. 97.0 C
2. 97.0 C
3. 97.0 C
4. 95.3 C
5. 95.3 C
6. 96.1 C
7. 95.8 C
8. 96.2 C
9. 95.7 C
10. 95.5 C
11. 95.9 C
12. 95.8 C
13. 95.7 C
14. 95.5 C (looks pretty stable)
15. 95.2 C
16. 95.7 C
17. 96.0 C
18. 96.1 C
19. 96.3 C
20. 96.2 C
21. 96.0 C
22. 95.8 C
23. 96.0 C

waited 30 mins and then repeated 19 more pulls: all temps between 94.9 and 96.1

Sunday afternoon same process different settings:

................1.2 bar................1.2 bar ....................1.4 bar.................1.5 bar


1.............91.9.....................92.1.........................92.0......................94.8
2........... 90.9.....................92.5.........................91.9......................95.5
3........... 92.0.....................92.8.........................92.5......................95.0
4........... 91.4.....................92.4.........................92.6......................94.6
5........... 90.1.....................92.1.........................92.3......................94.0
6........... 89.0.....................92.3.........................92.0......................93.1
7........... 88.8 ....................90.3.........................91.7......................92.4
8........... 88.6.....................90.2.........................90.7......................91.4
9........... 87.6.....................90.4.........................91.2......................90.8
10......... 88.5.....................89.5.........................90.5......................90.5
11......... 88.0.....................89.2.........................90.2......................90.5
12......... 88.2.....................88.6.........................89.6..................... 90.1
13......... ***......................89.6.........................90.3......................90.0
14......... ***......................*** .........................90.2......................90.3
15......... ***......................*** .........................90.4......................91.1
16......... ***......................*** .........................89.8......................90.5
17......... ***......................*** .........................89.8......................90.7
18......... ***......................*** .........................89.2......................90.4
19......... ***......................*** .........................89.2......................90.6
20......... ***......................*** .........................89.3......................90.7
21......... ***......................*** .........................89.1.....................***


So what do you think? The temps from Sat. night are great! Maybe less is on the current line so late at night? Anyway the variac comes this week, will try again with that and see if it makes any difference.

[Ken Fox]

Well, I doubt Jim Schulman will be coming by to PID my machine, so I continue on this journey, with the help of all of you.

Testing: (snipped)


So what do you think? The temps from Sat. night are great! Maybe less is on the current line so late at night? Anyway the variac comes this week, will try again with that and see if it makes any difference.

I'm not familiar with this testing setup

Honestly, I have no idea how to relate to these numbers.

Jim has visited me a couple of times to go skiing; he has an annual ski trip he does with a family member not far from here so he has added a visit here at the end of those trips. When we did the comparison between my old pstat controlled vibe machine and my newer (pre-pid) rotary machine we could not detect any difference in shot quality. When we did the PID study we found only one significant difference which was that the presence of the PID in the boiler reduced first (after flush) shot temperature variability by a smallish amount, but that smallish amount was enough to make the first shot temperature as consistent as later shots in a series. Normally, the machine does not get quite as temperature stable until the 4th or 5th shot in a series. Even so, I"m not sure that the benefits we observed would justify the effort of PIDing this machine, unless you also want to be able to easily adjust brew temperature which is a bit of a nuisance if you have to take the case off to get at the Pstat.

Still, with the machine running on the pstat, we did not observe inter shot temperature differences as broad as you report for some of these series. And, I don't understand at all the reported "brew pressure" settings. Using a PF manometer (which is what you need to measure brew pressure) you would get around 9 bar on a properly adjusted machine. An unadjusted vibe (not using the bypass valve) would produce pressures up to 14 bar or so. A rotary machine can produce nearly anything you want but anything brewed at below ~8 bar of brew pressure probably would not be considered to be "espresso."

ken

[SamuraiE]

Thanks for your comments, Ken. I am using a pf 20atm gauge to measure group pressure....set to 11 bar....sorry for the confusion.

I suppose the tests were to verify if the machine has a problem with temp drops over a serious of pulls...as my shots tend to get more and more sour as a sessions continues.

PID sounds nice....but with the outrageous prices of parts and service over here I doubt it would be possible.....unless I can convince Jim and his family to suddenly take up surfing and start making annual hops across the world in search of 6 ft. long righthanders.

[Ken Fox]

Thanks for your comments, Ken. I am using a pf 20atm gauge to measure group pressure....set to 11 bar....sorry for the confusion.

I suppose the tests were to verify if the machine has a problem with temp drops over a serious of pulls...as my shots tend to get more and more sour as a sessions continues.

PID sounds nice....but with the outrageous prices of parts and service over here I doubt it would be possible.....unless I can convince Jim and his family to suddenly take up surfing and start making annual hops across the world in search of 6 ft. long righthanders.

Well if you are set to 11 bar, you could consider adjusting the rotary pump to read 9 bar; it is a very easy adjustment and the owners manual shows where the screw is located that needs to be turned on the rotary pump. I have no idea if this will make your shots taste better, however.

If I were you I'd set my boiler temperature to around 1.1 bar maximum (or 1.05bar), with mimimum about 0.15 to 0.18 bar lower, depending on how well your pstat functions. If you don't like the end product than I'd be more likely to question the coffee you are using or the way the shots are being made than the equipment itself.

ken

[SamuraiE]

Hey Ken, I hear you...but I don't have a rotary pump, this is a S1, vibe pump, like your older machine. Shots are sour, thus the temp tests to see if it is a problem with the machine or with....................you know.............me.

[HB]

Pours were about 8 seconds which gave me 165 ml or 5.5 oz of water....didn't use a pf, just poured straight from the group right into the carton. Wanted to draw less water but didn't realize until too late that I was getting such a high volume. Pulled for about 8 seconds, waited 30 seconds then pulled again, and again, etc. (as per Dan's orders).

Sorry, the extra volume invalidates your results. Even so, what strikes me as unbelievable is that a BLAZING hot boiler setting of 1.6 bar doesn't produce flames pouring out the sides! The evaluation model I tested was set to 0.9 bar when it arrived and I tried settings ranging from 0.7 to 1.3 bar.

Shots are sour, thus the temp tests to see if it is a problem with the machine or with....................you know.............me.

If you and Ken agree to re-run the test I described above at the same setting (e.g., 0.9 bar) and compare, I would expect them to correlate within a couple degrees, even with the weakness of the "Styrofoam cup" technique. After that, the only remaining logical steps are (a) calling in the expensive service guy, (b) breaking out the tools on a witch hunt.

PS: Did you note the boiler pressure gauge behavior during the test? How long did it take to recover? Normally Junior's heating element will not click on after every extraction, but rather every second or third because of the huge HX.

[SamuraiE]

Dammit Dan! All that work for nothin'! Oh well....Yes, I am an idiot.

Let's see, I wrote to La Cimbali directly and Mr. Carlo Tambussi says the I should set the boiler to 1.2 bar and temp should be 70 C in the cup (geez, I knew that, thanks for nothing). But he does say that all Cimbali products have a standard length injector (so a least we can rule out that). Then he goes on to add that I should call my distributor (ain't got one).

By the way, the heating element did not kick on for every shot as you say, it kicked on every third or fourth.

OK, before I sell my 4-year-old son to pay for the home visit.....I will wait for the variac to arrive and give that a go.

If I try the test again at 90 ml, what temp would be ideal in my cup? 92 C?

Thanks

[Ken Fox]

Hey Ken, I hear you...but I don't have a rotary pump, this is a S1, vibe pump, like your older machine. Shots are sour, thus the temp tests to see if it is a problem with the machine or with....................you know.............me.

Cheer up, there isn't any clearly visible difference in the cup, vibe vs. rotary. The pressure on your vibe machine can be adjusted to produce a steady 9-ish bar. Most probably your machine lacks the silicone tubing needed to run from the nipple of the overpressure valve back into the pourover tank. I would not even try to suggest where you might buy silicone tubing in Japan, but I got mine from Mcmaster Carr in Los Angeles, and attached the tubing to the nipple of the valve with a twist tie. The valve is located just behind the front panel, so to get to it you need to take off both sides of the case and then the 4 screws that hold the front on; none of this is difficult. The OPV likely will need to be adjusted several times over a week in order to get it to adjust and to stay adjusted. Once it seats right, after a few tries, it should stay where you put it. If you aren't now using your OPV, most likely your machine is producing about 14 bar peak pressure with each shot. I don't know if that itself could be producing some of what you are tasting in your shots.

As to using a cup to test temperature, I think that method has so many limitations that I'd be reluctant to base anything on it. I think that if you set your machine to 1.1 bar maximum on the front panel gauge, then flush your group until it stops flashing (and then some), you will eliminate temperature as the cause of your problems. I still don't exactly understand what the problem is, but if there isn't anything obviously defective in your machine (and from what you have written I think there is not), then the fault lies in something else, such as the coffee you are using (which might just not be to your taste, who knows), elements of barista technique as apply to this commercial machine, or something else. This machine is more demanding when it comes to barista technique than your typical home or semi-commercial unit. When I don't distribute the grinds carefully in the PF, for example, there is no getting around it; a sink shot is coming. I am beginning to doubt the problem is in your machine, and perhaps you need to look elsewhere into elements of technique or the beans you are using. If you are using home roast, the fault could be in the roast level or how you get to that roast level. I myself have made every mistake in the book when it comes to roasting so I would not discount that possible cause, either.

ken

[SamuraiE]

Ken, how quickly you forget....or maybe you just help so many Jr. owners that we all begin to run together in your mind. You gave me all that great group pressure info (plastic tube, as well) several months back. I had the group pressure down to 9 bar and ran the plastic back into the tank for months....worked out great! But for the these tests I decided to raise the group pressure back up to 11 bar.

As for setting the pstat....it was at 0.9 for the longest time, but shots are cool and taste is sour, the more shots I pull the more sour they get, so I cranked the boiler back up to 1.6 bar with the hopes it would increase the temp and get the color of my crema darker than it is. The machine just doesn't seem to get hot enough. As for beans, I have a Hottop and roast Monkey Blend from SM, into 2nd crack. The roast always look great.....so have a hard time talking myself into all of this being bean related. I keep thinking the voltage in my old Japanese house just isn't giving Jr the juice he needs, thus cool, sour shots.

Of course we can't rule out user error.....pf distribution, tamp, etc.

Know you are up to speed on the problems without having to read through the entire thread. Thanks for your comments.

p.s. I wacked off the doser on my Cimbali grinder....it is now doserless....Japanese machine shop made a spout....will let you know how it works out.

Key

[Ken Fox]

Ken, how quickly you forget....or maybe you just help so many Jr. owners that we all begin to run together in your mind. You gave me all that great group pressure info (plastic tube, as well) several months back. I had the group pressure down to 9 bar and ran the plastic back into the tank for months....worked out great! But for the these tests I decided to raise the group pressure back up to 11 bar.

As for setting the pstat....it was at 0.9 for the longest time, but shots are cool and taste is sour, the more shots I pull the more sour they get, so I cranked the boiler back up to 1.6 bar with the hopes it would increase the temp and get the color of my crema darker than it is. The machine just doesn't seem to get hot enough. As for beans, I have a Hottop and roast Monkey Blend from SM, into 2nd crack. The roast always look great.....so have a hard time talking myself into all of this being bean related. I keep thinking the voltage in my old Japanese house just isn't giving Jr the juice he needs, thus cool, sour shots.

Of course we can't rule out user error.....pf distribution, tamp, etc.

Know you are up to speed on the problems without having to read through the entire thread. Thanks for your comments.

p.s. I wacked off the doser on my Cimbali grinder....it is now doserless....Japanese machine shop made a spout....will let you know how it works out.

Key

Hi Key,

For some reason I have gotten an undeserved reputation as some sort of Cimbali Junior guru; as a result I get frequent emails about Cimbalis (often from people who lurk in forums but never post), and I participate in a couple of forums plus a.c. Unless I know someone personally it becomes very difficult to remember who is who especially when a lot of the questions are similar. And, when you have people who use more than one name (e.g. "Key," "Samura," as but one example) then I really haven't a clue as to whom I'm talking to!

Temperature is temperature regardless of how it is achieved. If your local current is on the low side (and there is variation in household current), but your pstat shows a particular pressure (correllates with temperature), then the temperature you have is the temperature you have. If it takes the heating element in your boiler 25% longer to recover after frothing than mine takes, still, once you get to the temperature/pressure setpoint, you are at that setpoint. And, the thermal mass of a Cimbali Junior is such that you should be able to make a shot or two or three without the (potentially lower) house current having much impact. Sure, if you are making 100 shots per hour, you are going to notice it, but if we are talking about typical domestic usage, I really doubt you would notice it. Of course, you could have a level of current that would be SO LOW that it would damage the pump, but I don't think you have observed that.

Have you considered the possibility that you got a bad batch of green beans? I have received bad beans before without having changed suppliers. There was a celebrated incident of an online merchant selling Monsooned Malabar that was labelled as Malabar Gold. sh** happens. That would be the line of thought that I would pursue because I sincerely doubt that a variac will help you one way or the other.

Good luck!

ken

[HB]

If I try the test again at 90 ml, what temp would be ideal in my cup? 92 C?

Closer to 94C would be better. BTW, I agree with Ken's thinking that the blend or technique are at fault. When I start with an evaluation, I always use a standard blend (e.g., Intelligentsia's Black Cat) and bottomless portafilter to diagnose problems early on. You might switch to a pre-roasted blend from a reliable source to eliminate one more variable.

[SamuraiE]

Dan, Others, I got it! I did it! Espresso Nirvana! Tonight I got the temp bug figured out!

The Key (no pun intended) was Dan's advice of 94 C in the cup. I began really working all the angles on how to get that temp in the cup.

Using just a shot glass, a pf (without its basket) and a dig temp thingie, I started flushing away. Here is what I found out. Firstly, the variac didn't seem to play much of a role in any of this (for those keeping score at home that makes it: Dan- 1...Key- 0), my espresso machine repeatedly gave me a standard 88 C in my shot glass after long idle periods (anything over 3 mins down time). I next flushed 60 ml at 30 second intervals and watched the temp rise steadily until it reached 93.4 C after the 5th flush.

Next, I tried 60 ml flushes at 15 second intervals......after the 5 flush the temp reached 94.5 C in the shot glass.

So.....after long idle periods the machine is cool, and the flushes serve to warm it up (but you knew that already, right?). Lots of things to report:

It seems important to wait the 15 seconds after the 60 ml flush, when I tried pulling quicker than 15 seconds (say at 5 or 10 seconds after the 5th flush) the temp was always low again....around 92.5 C.

Also, it takes 5 flushes to bring the machine up to 94 C.....4 flushes won't do it (93.5 C).....only after 5 flushes.

So to pull a shot, 5 flushes of 60 ml at 15 second intervals....lock and load the machine and 15 seconds after flush #5 pull the real shot. It worked perfectly! NO SOURNESS AT ALL. The shot looked terrific out of the pf and tasted even better.

Remember this is a La Cimbali M21 S1 pourover model made to run on Japanese specs of 100 volts.

Finally, another big factor in all of this was that I finally finished my K.A.T. (Key Auto Tamper) this week. It is a Japanese drill press holder fitted with a stainless tamper and works a lot like the Macap Auto Tamp. It really shapes the puck beautifully and I can't but think that it played a huge role in creating tonight's honey-like pour.

Thanks so much Dan, everyone. I don't need that $250 home visit after all!

7 months after buy date I finally can drink a decent cup of espresso. I guess the lesson here is that every HX machine is different and has its own flush cycle.

Key