Unreliable Ulka pumps

Need help with equipment usage or want to share your latest discovery?
Posts: 614
Joined: Dec 03, 2012, 10:56 am

Postby Nate42 » Jun 03, 2013, 12:15 pm

My Anita has the fairly common Ulka EAX5 vibratory pump. This weekend it started stalling out on me during shots (flow drops to zero and pressure fluctuates). I'm all too familiar with this problem, and after verifying that my expansion valve is working properly and I don't have any electrical problems, it looks like once again the pump is wearing out. This will be my 4th one in as many years. I replaced it last 6 months ago, almost to the day. This is getting pretty frustrating. My next machine is DEFINITELY going to have a rotary pump, I'm not fooling around with these vibes any more.

I've turned down the pressure a bit (down to 8 bar during a shot now) and left the cover off to help dissipate heat, and that was enough to get it working again. I may eke another couple weeks out of it. But i've already ordered a replacement, I've been down this road before and it won't be long now before its completely unusable.

It seems like I've heard of people replacing their ulka pumps with fluid-o-techs. Anyone have any info on what is needed to make this happen? Are they really any more reliable?

Posts: 164
Joined: Jan 06, 2009, 12:15 pm

Postby Miami_AJ » Jun 03, 2013, 2:51 pm

Strange. I have an ULKA going on strong for 7+ years in my Expobar Pulser. I thought they were reasonably reliable but loud

customized for espresso aficionados
Sponsored by La Marzocco · Home - customized for espresso aficionados
Posts: 614
Joined: Dec 03, 2012, 10:56 am

Postby Nate42 » Jun 03, 2013, 3:17 pm

I think the 7+ years is the key. I think the quality of the Ulka pumps used to be better than it is now. If you look around on the forum you can find other similar complaints. The guys at Chris Coffee have told me that the longevity varies a lot, from 6 months to 6+ years. Unfortunately I seem to be stuck with the lower end of that number.

Chris coffee actual sent me this latest one that failed for free, since its predecessor lasted less than a year. They might well do it again but I didn't even ask them about it this time though. I figure its not their fault the pumps fail, and I don't feel like they owe me two freebies in a row.

User avatar
Posts: 420
Joined: May 30, 2007, 4:54 pm

Postby civ » Jun 03, 2013, 3:42 pm


I have seen similar posts here at HB (and other sites) quite a few times but nowhere have I seen a diagnostic. ie: what happened with the pump. I may have missed it.

From what I have seen, they are relatively simple artifacts, with a coil a couple of springs, 'O'Rings and only one moving part and I have the idea that many times it's just a question of taking it apart and giving it a thorough clean up.

Now, if the coil went south for some reason, you are probably looking at a new pump but it does not seem to me that the rest of the components can break down although one can be damaged if the pump is not connected properly.

And all of them can be very easily lost/misplaced if due care is not taken when disassembling it. =-/

Image courtesy of erics
For example, in the diagramme above, the little piece circled in red can be easily damaged by whatever goes into the threaded tube if it goes in too far.

Here's one post at HB:
Repairing a ULKA vibratory pump

Here's a site that comments on Ulkas:

Unfortunately, there are no repair kits for Ulkas.



Posts: 614
Joined: Dec 03, 2012, 10:56 am

Postby Nate42 » Jun 03, 2013, 4:02 pm

I have disassembled and cleaned my failed pumps before, using that very reference. Unfortunately I saw no obvious problems and cleaning did not help.

If anyone out there with a failed pump has a better idea of what specifically is failing, I would love to hear it.

User avatar
Randy G.
Posts: 3952
Joined: May 12, 2007, 11:54 am

Postby Randy G. » Jun 03, 2013, 5:36 pm

Your first post, relating that when allowed to rest for a while it would work again, might indicate overheating of the coil. Why your machine and not many others of the same model? Is there adequate ventilation around the machine? Is it in a location where it is drawing or recirculating hot air? Are there ventilation holes on the bottom of the machine to allow convection to draw in cool air? I am just spitballing here. Maybe mount it externally? Maybe replace it with an outboard rotary in a cabinet? Lever machine with manual boiler refill?
Espresso! My Espresso!

Posts: 45
Joined: Mar 01, 2010, 12:56 pm

Postby scrutinizer » Jun 03, 2013, 6:57 pm

Given the large number of machines out there using these pumps and relatively few complaints posted, I would question your conclusion that the repeated failures that you have experience are an indication these pumps are unreliable. My original ULKA pump (same model) lasted 6+ years and its same replacement (both from CC) is going on 2+ years and running strong. The 'norm' has previously been reported as approximately 5 years. Before blaming the equipment, I would be certain that all potential environmental variables that could contribute to early pump failure were fully explored.

One variable that seems to be frequently overlooked is inlet water temperature. Higher temperature water is less dense and causes the pump to work harder. I recall a previous posting from reputable source that these ULKA pumps are designed for pumping [approx...going from memory here] up to 75 degree water though I couldn't verify this w/ a quick search. Most tank espresso machines quickly heat the tank water to 90+ or even 100 degrees or more. Some tap water is also quite warm coming from the tap. Note that water that feels neutral to touch is probably in the upper 90+ degrees so tank water should feel cool to touch. An interesting test is to measure the tank temp and see where its at...you may be surprised at how hot it is. Over the past year, I have paid more attention to keeping my tank cool (even dropping a few ice cubes in when idling for hours) and the pump at least appears to sound less stressed when it operates (more solid/less raspy and labored).

Of course, what goes into the pump, continuous operating times, case heat etc are other variables that could shorten life. Going through pumps every 6 mo is rare (otherwise we would be hearing about it in much greater numbers) and so unfortunately the most likely answer is its your machine or how you are operating it or both. Before replacing the pump, you could try using 70 degree water and see if the performance improves.


exceptional coffee from exceptional relationships
Sponsored by Nossa Familia Coffee - exceptional coffee from exceptional relationships
Posts: 614
Joined: Dec 03, 2012, 10:56 am

Postby Nate42 » Jun 03, 2013, 7:02 pm

The pump has a klixon thermal fuse that is supposed to shut down when it overheats, and to my knowledge it has never triggered. The fuse isn't the thing that's failing because the first thing I always try is bypassing the klixon.

I doubt my environment temperature is a major issue, as I live in New Hampshire and my house is quite cool for much of the year. And yes I have had failures in the winter.

The Anita doesn't have any ventilation slats on the bottom, just at the top in the cup warmer area. My machine sits on a kitchen counter, with space to either side, so it has room to breathe. There is a cabinet overhanging it which I'm sure doesn't help, but that's not exactly an uncommon configuration. Since the water reservoir is sitting inside the hot housing heating up that doesn't help the pumps thermal situation, but of course every non-plumbed machine has the same issue. The pump is actually rated for water input temperatures no greater than 25C, and obviously any espresso machine with an internal reservoir is going to have higher temperatures than that. Still, many others are in the same boat, as virtually every vibe pump machine uses this same pump.

Issues with these pumps are not uncommon, but I do seem to have particularly bad luck with it. I don't guess I can rule out that something unique about my configuration is exaggerating the problem, but I don't know what it would be.

As to mounting a pump externally, that might well work, but I'd rather not go there, and frankly I shouldn't have to. I use the machine as designed, and it should work. Can't afford it at the moment, but I've got an itch for a double boiler machine, and I'll buy one of those (with a rotary pump, for sure) before I do any extensive modding of this machine.

On the upside, i've gotten pretty good at changing out pumps. :)

Posts: 614
Joined: Dec 03, 2012, 10:56 am

Postby Nate42 » Jun 03, 2013, 7:14 pm

scrutinizer wrote:One variable that seems to be frequently overlooked is inlet water temperature.


Before replacing the pump, you could try using 70 degree water and see if the performance improves.


I'm much more conscious of inlet water temperature these days, given that I've had so many failures and I've noticed the 25C rating. For example, I make it a point to ensure I have a full tank of fresh water when I do something that will involve a lot of pump stress like backflushing with detergent.

This most recent time when I was fiddling with it this weekend, I was always running it with cold water (from the fridge). It would still stall. I had to both reduce the pressure to 8 bar AND use cool water to get quasi reliable performance.

I suppose when I get my new pump I could start refilling the machine with cold water prior to each use, but frankly that sounds like a pretty big pain and I've never heard that this is something that is necessary. Still, its probably less of a pain than changing pumps ever 6 months.

Again, I don't feel like I'm doing anything unusual as far as my usage of the machine.

Posts: 23
Joined: Nov 28, 2011, 9:09 pm

Postby Metatron » Jun 03, 2013, 7:49 pm

It may be the switch that is activating the pump has a problem. It may have high resistance contacts which are dropping the voltage to the pump. I think this may increase the amperage drawn by the pump thereby producing less power but producing more heat. Also, if the contacts are arcing, the pump will probably get very confused as to whether it should be pulling or pushing.

If the switch is solid state and is faulty, it can get into a partially on state, which is not good.

It seems you have covered most of the bases but it may be something like the switch. Or, maybe it is haunted in which case call the Ghostbusters.