Ok, I picked up a used Bezzera BZ99 a few weeks ago, and it'd worked fine... I'd noticed that there was this weird leak coming from part of the piping that goes from the pump to the boiler, but it seemed minor and it didn't bother me too much. So, this evening, I decided (having gotten some citric acid) that I was going to descale the machine as it probably hadn't been descaled before I'd gotten it (the machine was a commercial pull; it saw ~7 years of restaurant use).

So, I opened it up (which I'd done before; Bezzera makes it remarkably easy) so I'd be able to override the boiler autofill and such while I descaled. And, foolishly, I decided to do a little tinkering with my leaky piece of piping. I localized the leak to a T-junction in the aforementioned water line and I tried to tighten it using my trusty multitool (I'm rather short on tools; the multitool gets used for nearly everything)... and managed to snap one of the portions of the T-junction (the top one, which goes out to a overflow (?) line back to the water tank. There's about 1 thread on each side of the break... and this is in solid brass...

As I can't find anyone in the US who sells the parts in question, I decided to take a chance and order a couple pieces which ought to work from McMaster-Carr. At best, they'll work - at worst, I'm out $20 or so for little bits of brass.

At the moment, I'm infuriated with myself for managing to break the parts in question and disgusted that little brass parts, with walls ~3+mm thick can break like that when manipulated by hand tools.

If anyone's got good ideas and/or knows of a shop in the US who sells Bezzera parts, I'm all ears.

To illustrate the point, here's the diagram from the BZ99 manual and the parts in question...

You're seeing Assembly #14 and #7... in very, very bad shape.

I had a similar thing happen to me but it was just a water line intake elbow that I was able to just eliminate and use a brass part I found at Home Depot. It was the same Serto labeled part which might be a brand but I searched all over and didn't find the exact same part. Espresso Parts NW has a lot of brass fittings that they sell, but have you tried looking for Pasquini parts? I believe the Bezzera should be about the same internally as the Livia 90, and maybe there is a source for Pasquini parts. It's frustrating that the parts break and not only that but they're sometimes a really hard to find BSP thread. I checked Pasquini's web site and they sell some Livia parts but I didn't see this one shown, although they may have it if you contact them. It looks like 1st-Line is a Pasquini dealer that might also be able to help.

Quoting my friend Angelo from Espresso Parts Source, the only way to avoid ever breaking something on an espresso machine is to never work on them. Screw ups happen and I've certainly had my share.

What I would do if I were you is to forget about trying to match the exact parts that are broken but rather look at what the function is of the broken parts, and try to replace the functionality. What is being connected to what, and what function does this set of parts serve?

If, as I suspect, they merely connect two things and don't otherwise do anything (such as serving as an overpressure valve), then have a look at what is at both ends and just find something that will bridge the gap and that will work.

There are probably 8 to 10 different ways you could fix this, all of which will work, and which will probably require between $5 and $15 worth of plumbing parts.

I finally got sufficiently frustrated with just this sort of problem that I ordered a set of just about every possible connector and fitting from Angelo, and whenever I need to fix a screw up, I can go into my parts drawer and find something that can be cobbled together to work, and the machines are back functioning, as good as new.

ken

Went out and raided my local hardware store (snort; more like a yuppie accessory store with tools, etc in the back) and picked up a vise grip (which I needed). I've now disassembled the two components shown in the picture in my first post, and it looks like there was just enough threading to let things reseat... I'll post an update shortly.

Ha! There's still a minor leak (1 drip per cycle or so); basically where it was before. I'm not overly bothered by that and I've put some absorbent material beneath it to catch the drips.

Now, its heating up and I'm going to have my morning coffee.

Espresso production was successful, albeit slightly leaky. I'd also grabbed some "epoxy for stupendous bad-asses" (aka all-conditions epoxy) and I slathered a bit on the leaky joint (on the outside) - I do stuff like that at work all the time with lesser epoxies, and it works fine.

So, I'm going to leave the machine open to dry and I'll check on it when I get back from classes, etc at 4pm.

Make another trip to your hardware store and buy a tube of blue LocTite. Carefully disassemble the leaky connection, dry and apply the LocTite. Reassemble, wait 24 hours... You'll have a leak free machine.

John

john_K wrote:Make another trip to your hardware store and buy a tube of blue LocTite.

What about just using teflon tape?
It has always seemed to work on non-compression brass connections for me.

Susan

Eh; I went with what I had.

In regard to teflon tape, there just isn't enough threading left to get it to hold...

SJM wrote:What about just using teflon tape?
It has always seemed to work on non-compression brass connections for me.

Susan

Blue Loctite will work (sometimes) when teflon tape doesn't. You can try teflon tape, but if it doesn't work, try the Loctite.

My friends Angelo and Michael from espresso parts source, both of whom used to work as repair and design techs, have told me that blue loctite is used by professionals when they make repairs on commercial machines.

ken

It could be that the PO found the leak and over-tightened it, causing a crack in the brass that you only completed in trying to solve the problem. It appears that the assembly operates as a one-way check valve system, so it does need to be configured that way if you ae looking for almost-fit parts.

beanmuncher wrote:At the moment, I'm infuriated with myself for managing to break the parts in question and disgusted that little brass parts, with walls ~3+mm thick can break like that when manipulated by hand tools.

Awww.. don't feel bad. Back in 1970 I worked part-time as a tire-changer at Montgomery Wards in the San Fernando Valley when I was going to college. A fellow came in and requested that we not use the air gun on his lug nuts on his sports car. I got the 4-way wrench out (we didn't use torque wrenches back then- it was the days when cars were made out of real metal). Without trying, I easily twisted two studs off by hand. It was the only time that happened in the 15 months I worked there.

use green loctite, or, even better, just go ahead and solder the parts together.

at the moment, I'm just waiting for the epoxy to fully dry... it looks like it is going to work (I do stuff like this all the time at my lab)

if it doesn't work, i'll root around my junk machines to see if i can come up with something. i ought to have a similar tee someplace, and the other part appears to be a checkvalve, but with female threads on one end instead of the normal male/male.

Thanks! Hopefully, the epoxy will work, and failing that, the parts I have coming from McMaster-Carr might...

I just checked on the epoxy and once it fully hardens (read: sometime tomorrow) I should be good to go... the stuff hardens rock-hard.

barry wrote:if it doesn't work, i'll root around my junk machines to see if i can come up with something. i ought to have a similar tee someplace, and the other part appears to be a checkvalve, but with female threads on one end instead of the normal male/male.

This is oh-so-typical of Barry, the patron saint of home baristas . . . .

by the way, what is green Loctite? I've never noticed it but then, the couple of times when I used Blue loctite it worked fine, so I didn't need to resort to the "nuclear option" . . . .

It came up earlier and I found a good summary in the catalog (see pages 4-5):

• Loctite 242 Blue Threadlocker is an all-purpose, medium strength threadlocker for use on nut and bolt applications from 1/4" to 3/4" (6mm to 20mm).
• Loctite 262 Red Threadlocker is a permanent, high strength threadlocker for use on fasteners up to 3/4" (20mm).
• Loctite 290 Green Penetrating Threadlocker is a low viscosity threadlocker designed for use after assembly. The threadlocker "wicks" along the threads of assembled fasteners and locks the fasteners in place.

Ken Fox wrote:by the way, what is green Loctite? I've never noticed it but then, the couple of times when I used Blue loctite it worked fine, so I didn't need to resort to the "nuclear option" . . . .

Loctite 290


It is a wicking acrylic threadlocker, which happens to also form a pretty good sealant for small holes/imperfections. Great stuff.

Well, the machine is heating up now (it has been sitting out on my kitchen table , splayed open, for the last day and a half) and I felt around the epoxy when the pump kicked in, and I didn't feel any leaks. Provided that holds up, I'm not overly worried.

Now, to think about descaling... which is how this whole bloody mess started. Considering how the BZ99 is put together, you need to descale with the machine the way I've currently got it (e.g. minus the sides, rear and top panels), so I'll probably do that later today. I've got a concentrated citric acid solution (40g of 99% pure citric in 10oz of water); I'll dilute that out using the water tank in around 2L of water, pull it into the boiler and roll from there. Overriding the boiler probe is trivially easy... either pop the connection spade, or take a pair of pliers and extend it however much you want.

I figure I'll use the standard descaling for HX instructions - but does anyone have any other ideas?

Also; just got the parts I ordered from McMaster... they were cheap enough, but paying for Saturday delivery was damn stupid ($28 ?!?). Looks like they would have worked, if needed.