Cimbali Jr = sour everytime, why?

[SamuraiE]

Dan and others, any help here would be great! I've had my setup, Jr and Jr grinder for 7 months now. I think I have come to understand both machines and how to construct a nice drink.

Since day one all my espresso shots have been sour, never bitter, never once. I played with the settings after the first month and brought boil pressure down to 0.9 bar and same for group pressure. Grind is good, always dose same amount, standard Cimbali double basket filled to top....north/south finger sweep, always tamp the same. Shots pull great 60ml in 30 seconds. Crema 50%-70% at first then settles to about 15ml. Color of shot is fine (dark with specks) sometimes goes blond a little early. Got my bottomless pf last week just to check if my tamp, etc. might have been the problem but it wasn't....pulls are centered and no sprays.

Yesterday I set boiler way up to 1.4 bar just to see if I could burn a bitter shot and moved group pressure up as well. Even without a cooling flush and letting the machine set for 2 hours the shot came out sour again. So now I am thinking this has to be some kind of voltage problem in my house. Perhaps the machine is set to run on 110volts and is only getting a floating 100-110 volts? The folks in France who I bought from had La Cimbali make this machine to fit with Japanese specs and told me it was set to go at 100volts....but come on, over 500 shots and every one sour, what gives?

Key

[SamuraiE]

Should add that I have tried many different kinds of beans and have a Hottop and roast at home with good results so this isn't a bean issue.

Key

[HB]

500 shots, all sour? You have a lot of patience!

Unless the extraction is too fast, I would blame the temperature for sourness. Junior will spit plenty of steam and boiling water with a boiler setting of 1.4 bar after being idle for two hours (mine was set at 0.9 bar). If that's not the case, I suspect a mechanical problem. Have you had a service technician take a look at it? Did you say this machine is new? I ask because we all learned a lesson about scale in Sputtering e61 & HX scale build-up - Cured!

PS: As I mentioned in the review, although the heat exchanger contains super-hot water, the group itself idles below brew temperature. It takes a few flushes / shots to warm up. Unless you take extra steps to warm up the group (see How I Learned to Love HXs in the review), the first shot or two will tend to be sour.

PSS: I don't believe this is possible, but let me offer some additional wild speculation... maybe the water level is very very low and only the tip of the heat exchanger in touching the water. The water level sensor is "f" in the picture below. It is at the lowest point when vertical and the highest when horizontal.

[SamuraiE]

Dan, that was quick! Let me give you a few more details just to see if they mean anything to you. This is a brand new machine and under the drip pan it says:

La Cimbali M21 Jr. S1
100 ac (guess they really did make one especially to match these Japanese specs)
50/60 Hz
Boiler 1: 133 C-0.2 Mpa (2 bar)
Boiler 2: 133 C-1,2 MPa (12 bar)

Are they saying that the boiler should be set at 2 bar? That seems very, very high to me. And why Boiler 1 and 2....I thought this machine has only 1 boiler?

Finally, the boiler fill is set exactly like the one in your photo.

500 shots all sour, but they range in degree of sourness...some quite sour others not to bad....also milk drinks taste fine even with the sour espresso base and those are what we tend to drink the most of anyway. Perhaps I should write to La Cimbali and ask them what they think. Also, since I refused to pay the $8,000 price tag that the Japanese rep in Fukuoka was asking I doubt very much that his service tech the next prefecture over will want to have anything to do with my machine.....alas, I am on my own.

Key

p.s. read you article on the N.C. contest over at CG....very well done as always!

[HB]

I can't comment on the labels as I didn't bother looking at them closely (clearly Junior doesn't have two boilers and the rest of the information isn't pertinent).

Back to the sour shots: I would like a little more detail about the temperature. For starters, is the cooling flush sputtering similar to the video for the Elektra A3? Junior doesn't flash quite as loudly, but you should see a very distinct transition from the "water dance" to a smooth flow. Assuming that's as expected, do you follow the basic steps outlined in the writeup for warming the group and flushing before the extraction? My line of questions are towards figuring out if this is a mechanical issue (e.g., the HX and group aren't heating enough for some reason) or usage issue (e.g., flushing too much).

It would be very nice if you had a thermocouple thermometer to check the in-basket temperature; if that's not possible, the Styrofoam cup technique is worth trying. It shouldn't be hard to measure around 198F if the flush is correct.

PS: I don't think this is at fault, but it's worth knowing that raising the boiler's water level will increase the group temperature.

[bobroseman]

Until you figure it out, you might try letting the first 20 mL go to waste and just drinking the last 40mL. See if the sour taste is mitigated.

It's just a stop gap 'til you sort out the problem but I think it will work.

Bob

[AndyS]

clearly Junior doesn't have two boilers

1. Service boiler
2. Heat exchanger "boiler."

The heat exchanger has its own (higher) pressure rating.

[SamuraiE]

Thanks for all the good info! Here are my test results, fresh off the wire.

Cut a 1000ml milk carton down a little and stuffed it into a large coffee mug. Using tape, molded the carton to the standard Cimbali pf (with spout removed) and inserted standard double basket (no coffee just ran the shots straight through the double basket and into the carton). Punched small hole through side of carton and inserted my digital thermometer. In other words, pretty much no air entered the holding vessel. According to the standard guide temp should be within 190-204 F or 87.78-95.59 C

So here goes: Machine warmed up 1.5 hours and pulled about 5 or 6 flushes of 5 oz over a 20 min period.

Bar set to 1.3 and 5.5 oz flush before each shot about 2 mins between pulls

1. 87.5 C
2. 89.8 C
3. 88.0 C
4. 89.3 C
5. 91.4 C no flush before this pull just to see what would happen
6. 90.9 C
7. 89.9 C 4 oz flush before this shot
8. 93.6 C no flush and left machine idle for 5 mins (hottest of the session)

changed bar to 1.1 just wanted to bring it down and see what happens...much shorter flush as water dance ended more quickly (this had been the setting for the machine for the last 3 months or so)

9. 84.7 C 10 oz flush before this shot to stabilize the bar pressure (so should probably discount it)
10. 89.5 C this with 4 oz flush
11. 90.0 C also with 4 oz flush
12. 90.3 C 10 min wait between pull/5 oz cooling flush

OK, so what do you think? Temp seems to be within the guidelines albeit on the lower side.
This machine from day one really seems to need a fairly long flush (5 oz) just to get the water calm and smooth out the dance.

Should I crank the boiler way up to 2.2 bar or something crazy like that just to see what happens? Should I go back to 1.3 bar and not flush the machine in hopes of getting results like pull #8 (93.6 C)?

Just to see what would happen I just ran downstairs and pulled a shot in the milk carton to see what the temp is (carton is not preheated so that should take the temp down somewhat).

Liquid Amber espresso beans....just roasted 3 days ago (72 hours rest) with my Hottop, well into second crack. Cimbali double basket full to top...north/south...tamp and 30 sec pour after a 4 oz flush. Nice color and lots of crema.

Temp 78 C.

Taste, first sip a little burnt....then sour. Not undrinkable sour...but sour, again.

[HB]

OK, so what do you think? Temp seems to be within the guidelines albeit on the lower side. This machine from day one really seems to need a fairly long flush (5 oz) just to get the water calm and smooth out the dance.

Yes, I think we can safely eliminate mechanical issues. It looks like you are overflushing the group. I made the same error during the early days of the review due to my experience with prosumer E61s; the Cimbali Junior still needs to be flushed, but not nearly the same volume or frequency as a prosumer E61. For those unfamiliar with Junior's group, below is a nice diagram from the owner's manual:


Cimbali Junior hydraulics diagram

The heat exchanger is the cigar-shaped tube resting in the boiler water. The thin tubing inside it is the injector coming from the pump. This "dead end" heat exchanger is different than your typical prosumer E61, which is usually nothing more than a section of copper tubing passing from one end of the boiler to another:


E61 group hydraulics diagram

The grouphead is heated by the thermosyphon bringing heat from the boiler (red / blue arrows). Junior's grouphead is directly attached to the boiler by a heavy "trunk" and picks up its heat by conduction. The water in the heat exchanger also circulates heat to it (hotter water rises, cooler water falls). There are practical consequences of these two designs for the barista. The one that's most likely in play for your situation is Junior's "group memory." Because of the heavy trunk, large insulated boiler and heat exchanger, and beefy grouphead, Junior's temperatures are phenomenally stable under a hard load, which is what the machine was designed for. Junior's temperature management is tricky if it is being used intermittently, and worse yet, if you inadvertently overflush a few times, he'll remember. Other machines I've used don't have as a long a thermal memory -- three minutes and most forgive all mistakes. Not Junior.

In the article I described how I programmed the dosing keys to make flushing easier (Big Flush = 120ml, Double = 90ml, Mini Flush = 60ml, Spritz = 30ml). I only used Big Flush once in the morning and the rest are Mini Flushes. Your temperature trials were all preceded by Big Flush+ and likely accumulated some downward thermal memory. As I mentioned earlier, I got caught by the same error and it was Ken Fox and Jim Schulman who pointed me in the right direction, which begins by forgetting what you know or have read about using prosumer E61s -- applying the same regime to Junior will result in consistently sour shots.

[malachi]

Agree with Dan but would also love to hear more about the coffee you're using.

[SamuraiE]

Dan, I swear I must have read your review of this machine 15 times (along with all of the Cimbali King, K. Fox's stuff...nice on CG, nasty on Alt.com if you tempt him) and it was a big reason I went with Jr. in the first place. I understand your comments on the length of flush yet just want to point out that if my machine is idle for even the shortest amount of time the water will just explode out of the group with a sound that would wake the dead. The smallest flush you mention in your review does not even begin to slow down the jig the water is doing, nor the steamy sounds of water on fire (Can water be on fire? It was back when I was a kid in the 70's, we could drive up to Cleveland and watch Lake Erie ablaze).

Anyway, it usually takes at least 4oz to get anything that resembles a smooth flow of water out of the group. So, I guess what you are saying is to not flush 4oz and just go ahead and pull the shot with all that steam and splash and hope it keeps the group up to temp. Do you think boiler setting is going to make any difference?

As for the beans I am using, I ordered quite a load of greens from SM and use Mr. Monkey, Liquid Amber, SM Classic espresso blend and have even used lighter roasted beans one would not normally use for espresso (think most of the usual suspects from Africa and S. America) but rather a popular selection for use in a press pot or vacuum setup. Anyway, they always come out on the sour side....yet taste fine with any of the other machines I have in the house (basic paper filter, vacuum, etc).

Oh well, my cappas are always great. Jr. foams like a king. Guess I will try to flush less and.........um.....flush less?

OK, I am listening and no suggestion is too stupid so throw me a few more bones.

[HB]

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).

[malachi]

Lighter roasted coffees are more likely to contribute to your sour espresso as they tend to require higher brew temp. This can get even more noticable if your roast time is very short and your roast degree is light (ie. incorrect profile).

Suggest staying below 1BAR - this will help.

[SamuraiE]

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.

OK, Dan, gang...I am off to tame the beast! I will set the bar back to 0.9 and flush less. And when the nuclear flash (spraying water) hits, I will remember the words from my 2nd grade teacher: "Duck and cover!"

thanks all, Key

[k7qz]

I've been following this thread with some interest as I've been using a Junior user now for several weeks and he is clearly a different beast than the E61 "Pro" machine I used to have.

One thing that really stands out to me so far is the brewhead temperature stability of this unit. I don't know how many shots one would have to pull in an hour to overwhelm Junior but I don't think it would be an easy task!

I've also read Dan's Junior article several times and I agree it was very well done (like his others!). The recommended cooling routine suggested there was:

"If it's the very first time of the day (really, really idle), this flush raises the group to temperature, but a few extra flushes help get the big hunk of metal components called the brew group all singing to the same tune. To simulate the heating that naturally occurs over the course of the first half-dozen shots, follow Big Flush with Double three minutes later, and then three Mini Flushes at one minute intervals.

It's a few extra steps that really improves the first shot of the day. To summarize the early-morning startup routine: Big Flush, pause three minutes, Double (flush), pause one minute, Mini Flush, pause one minute, Mini Flush, pause one minute, Mini Flush, pull your shot." ... snip

I wonder, given Junior's memory, if one might start with a Double flush rather than the Big Flush? I don't have a temp gauge to test this hypothesis.

Dan, any thoughts?

-Mike

[HB]

I wonder, given Junior's memory, if one might start with a Double flush rather than the Big Flush? I don't have a temp gauge to test this hypothesis.

Unfortunately I don't have Junior any longer to test. It would have been nice to have a thermofilter to verify your hypothesis, but I'm certain Junior needs a big kick in the pants to get out of bed. The Big Flush serves this purpose. I documented similar behavior for the Elektra A3, but since it doesn't have volumetric dosing, I simply add three extra seconds of flushing first time out.

[SamuraiE]

Dan and anyone else who was kind enough to throw some advice my way.....thanks......I need a little more. If you recall Jr. gives me the sourpuss each and every time. Well on the last thread we covered flushing and warming up the group. I even went to the extreme of jacking the bar pressure as high as it would go to see if that would give me a bitter shot. It did, a little, I suppose.

Well after reducing the flush between pulls the shots taste better but usually get more sour the longer the session goes (first shot is always the best).

So here is my question: Is it possible that my machine just isn't getting the power it needs to fully reach proper brew temp? I'm thinking this is what is going on. The machine was built to match Japanese specs (100v) but the voltage in Japanese homes really spikes and drops sometimes well below 100v. Could it be that a line regulator (variac) would solve this problem? I talked it over with a fellow Jgeek and he suggested I ask everyone here at HB.

Would a lack of constant 100v of current cause the machine to not reach proper temp? As I mentioned above, just for the hey of it, I jacked up the bar pressure as high as it would go (1.6 bar) and without flushing before the shot managed to pull a really full-bodied, dark, tasty, very hot shot. Much darker than any pull at a lower setting. However, sure enough the next shot was not nearly as hot and had sour tones (no flush at all between the 2 pulls except for a pf wiggle immediately after the first shot).

Ideas?

Key

[HB]

So here is my question: Is it possible that my machine just isn't getting the power it needs to fully reach proper brew temp?

At times like this I wish you had a thermofilter so we could compare data with confidence. If your boiler pressure gauge indicates the proper pressure (e.g., 0.9 bar), then the lower voltage's only effect temperature-wise would be slower recovery. Reviewing this thread, I don't see any mention of the brew pressure. Have you checked that too (around 9 bar)?

I'm not convinced that there's a mechnical issue at fault, however a shortened HX injector would produce overall lower brew temperatures. A little background on this... When I first evaluated the prototype Andreja Premium, I noticed it pulled consistently cooler shots than other similar prosumer E61 machines at the same boiler setting. It also behaved much like you described - if I cranked the boiler setting way up to 1.6 bar, the first shot was close to the target temperature, but fell off rapidly in subsequent shots. I asked Quick Mill about it and they explained that the prototype was constructed using an injector length for the desired temperature range for the German market, whose brew temperature preferences run slightly lower than in the US. The Buyer's Guide was based off the production model destined for the US and behaved like the other prosumer E61s described in How I Stopped Worrying and Learned to Love HXs. It is possible that Cimbali similarly tunes their machine's natural temperature range for different markets; your local repair technician would know for certain.

[SamuraiE]

Dan, I have a group pressure gauge I bought from coffeeparts and I adjusted the group pressure from the preset 12 bar down to 9 bar. So I think we can be pretty sure that all the bar settings are correct.

As for your comments on injector length, I have to admit I don't know what you are talking about. Is this something I can adjust? If so, how do I reach it? Just looking at the diagram of my machine a few posts above I see 'Injector' but can't really find it in the pic.

Should I reset the group pressure back up to its original setting? Or maybe just start smacking the machine around a little, you know, get real tough with him....start dropping hints about Yakuza gang members on the way over to have a little chat with him, etc.

I am sure all this is just a simple adjustment away from getting the machine hotter and pouring out great shots....just I'll be damned if I can figure it out.

Don't know any techs around here....since I bought direct from Europe and not from the Japanese rep I doubt these islanders will be willing to come and have a look see at my machine....but I will call and endure their wrath just the same.

Key

[HB]

As for your comments on injector length, I have to admit I don't know what you are talking about. Is this something I can adjust? If so, how do I reach it? Just looking at the diagram of my machine a few posts above I see 'Injector' but can't really find it in the pic.

It's the thin plastic tube that runs down the middle of the heat exchanger and isn't meant to be adjusted. Owen's Cimbuini surgery, Jan.24, 2004 has an excellent picture of it. BTW, though it has nothing to do with your problem, his commentary on the replacement of a damaged heat exchanger is hilarious and should be read from the beginning starting from "Jan 21" to fully appreciate.

I am sure all this is just a simple adjustment away from getting the machine hotter and pouring out great shots....just I'll be damned if I can figure it out.

I'm not so sure because the adhoc data you collected wasn't remarkably cool given the weaknesses of the measurement technique (low 90s) and potentially excessive flushing. My comment about the injector is grasping at straws. See if you can curry favor with a local repair technician and I'll ask one of the technicians at Chris' Coffee to read this thread and hopefully offer some new ideas.