Sage/Breville Barista Touch reverse engineering (began with a heating problem)

Equipment doesn't work? Troubleshooting? If you're handy, members can help.
old_bear

#1: Post by old_bear »

Hi,

I have a long story of frustration with Barista Touch: suddenly, it stopped heating water, espresso shots came out cold, same happened to "hot water" and steam.
Over many experiments, it was noticed that sometimes the temperature goes up, then quickly goes down without reaching normal.
Investigation inside revealed no signs of trouble, no scaling (and the water was either very soft local one, or bottled), nothing. Service menu, however, showed heater shortening and E-Fast trigger as "1".

The only service I found in Norway is inadequate: they updated the firmware, replaced the solenoid group, sent it back and charged me a fortune.
Happily, they also sold me a heater assembly later, so that I could try to rule out thermal sensor or the coil.
Now, after replacing the heater, it felt well. I had stable temperature, and managed to pull some reasonable shots.
And then again: temperature goes up, then down, like thermal sensor gets mad.
"E-fast trigger" counter in the service menu still increases sometimes, following an attempt to use the heater but no other errors.

I am getting desperate, but still too stubborn to give up and throw it away.
Will appreciate any hints as for what to look at.

So far it seems that (considering the heater was replaced), "E-fast" is to blame. Apparently, the sensor on the heater is for NTC. But where is the overheat sensor then?

WWWired

#2: Post by WWWired »

Hi there old_bear :)

No need to throw away your valuable investment. Your story also underscores the importance of ensuring that you have a firm diagnosis of the issue from your trusted repair folks. If they aren't 100% sure, ask them if they can find the issue with certainty before just beginning to swap out parts (which can indeed fix a problem but only if the component in question is the source of the fault).

Can you take some photos of your machine (to identify the exact version/model) along with some pictures of the error codes displayed possibly?

Your Barista Touch is a brilliant machine that just needs a little more investigating to determine the exact source of the fault. There are many very experienced folks here at Home-Barista, and your machine's source of this problem will be solved in no time as people comment and add insight :)

For now, here's a picture of the "ThermoJet Heating System in both Breville Barista Pro and Barista Touch"


. . . and a picture of an actual Thermojet assembly from a Home-Barista posting by matebence96 . . .

old_bear (original poster)

#3: Post by old_bear (original poster) »

WWWired wrote:Hi there old_bear :)
Thank you for the reassuring response!
Unfortunately, the way most workshops work here (in my experience) is really random parts replacement and quick sending of bills. No one will take a risk of investigating, it's the customer decision to replace something - and if you agree, it may help, but if you don't, you just pay a little fortune for them to even look at it, and get it back broken. I made a mistake of not getting into a hassle of sending it to a service in some other land.

Yesterday I opened it again, disassembled the Thermojet (the new one this time) again and looked at the overheat switch:


What I noticed is that it was inclined very much towards the front of the machine, making the wide part in more tight contact to the coil plate.
Now, that happened due to the wire between the switch and the main contact group: it pulled it.
So I adjusted the wire a bit, and assembled it back, and it got "better": the average temperature got much higher, I did not see "E-fast trigger" (ER18) at least in a number of experiments, but overall the shots are still cold.

Since the flow is stable at around 16-17 ml/s (according to the service menu), the slow pump is hardly the reason, and I am now thinking that the trouble is somewhere around the overheat switch.

I will think on it a bit, factory reset the machine, look again, and then collect some pictures and maybe make a video of the numbers on the screen in "debug" mode.
In the meanwhile, any ideas inspired by the recent discoveries are very welcome.

WWWired

#4: Post by WWWired »

Brilliant old_bear :) . . . already your post is going to help many who are having similar issues and as you progress, this will provide increasing insight and any future video and/or photos you post will definitely give many here with experience and expertise some information to offer ideas and solutions as well :) Fantastic photo by the way! Very awesome! :)

old_bear (original poster)

#5: Post by old_bear (original poster) »

So, somehow I managed to make it work. By that time I was out of ideas apart from some bubble.
Still, I am sharing some sketches from the reverse engineering efforts.

Please feel free to contact me if anyone gets trouble with Barista Touch - I might be able to remember more.




old_bear (original poster)

#6: Post by old_bear (original poster) »

Since it was still unstable, I kept molesting the device.

The observation was the following.

In the service menu, the temperature seems trustworthy, but there is also some "E-fast" value. When the machine behaves well, it is somewhere below 1500 at rest, goes up to 1800-2500 in operation. When it feels bad, it jumps to 5500-6000 even at rest, and if it happens in operation, the temperature drops down.


When E-fast gets mad, sinister quiet buzzing sounds Deep Within.
Upon deep penetration, it was revealed that this buzzing source is somewhere near the upper corner of the mainboard. And there is a component there that looks indecent, marked ISO6 and pointed with a red circle on the image. Note: real upper corner is in the bottom of the image, because it is installed upside down in the machine.
It is called "EL 3063" and it is a very fancy semiconductor light bulb.



The blast crater is around the first leg, but it does not really say anything.



Now, I can imagine that floating state of this key results in one (third) solenoid not working sometimes (without system knowing about it, or the solenoid being mechanically flawed). That would not explain why hot water fails in all modes (since it is the last solenoid in the row), but floating faeces (somehow the word "shít" gets scrambled) in the control line may also mean something.

I ordered a seemingly compatible component and intend to change it and see.

ira
Team HB

#7: Post by ira »

I'm going to guess the defective part is marked ISO8. You didn't say which pin was blown, but as that's not a part that should fail that way, I might guess that something else died first, possibly TA6. But nice troubleshooting and I hope it works again soon.

old_bear (original poster)

#8: Post by old_bear (original poster) »

Oh! And by the way. The triangle shaped device I suspected to be related to the "E-fast" thing, is not that.
It is a simple one use thermal fuse, simple as a brick.
It either conducts current, or gets overheated and never does anything again, waiting to be replaced.

old_bear (original poster)

#9: Post by old_bear (original poster) »

ira wrote:I'm going to guess the defective part is marked ISO8. You didn't say which pin was blown, but as that's not a part that should fail that way, I might guess that something else died first, possibly TA6. But nice troubleshooting and I hope it works again soon.
It is hard to notice, but there is a deposit of a glass like black mass on the corner of ISO6, marked with a red circle. Updated the comment to include a mention of that.
My electronics advisor spotted it and suggested that some power artefact could have led to that.

And I agree that this seems strange, but nothing else stands out after thorough scrutiny. Very hard to make good images quickly, sorry for that.

ira
Team HB

#10: Post by ira »

Ah, I see now, wrong corner. So same thought, I'd also look at what it's driving.