Dead Auber PID - replacement options

[Korbah]

Hi all,


Sadly my Auber PID (SYL 1512B) died in my Rancilio Silvia. There's a burnt smell coming from the unit and there's a loose electrical component rattling around - clearly something within the PID unit has died an unhappy death. Frustratingly it's 13mths old - 1mth out of warranty.

I have the steam control option which has worked well.


One option is to purchase another PID from the company I originally bought from - at ~$90 + shipping.

Another option is to get a XMT7100 from china - delivered within a week for $30. The issue I have with this option is I don't know whether it'll work with the steam control like the Auber did. I also don't know the correct settings to program into the unit one I install it.

Thoughts?

Thanks in advance!

[Dimal]

G'day mate...

I guess the easiest option would be to purchase an identical replacement so it will just 'drop in'.
Also worth getting in touch with Auber I think and explain what's happened. A reputable company would surely try to assist given the 'infant mortality' nature of the failure.

Full on commercial/industrial grade controllers with similar functionality to the Auber will cost considerably more and as you've hinted at, will probably require setting up from scratch, and maybe even different configuration with respect to wiring/connections.

The cheap and cheerful Chinese units are a bit of a gamble, quite apart from simpler operation, you will most likely have to come up with an alternative Steam Generation strategy. Not that difficult to do but would require specialist knowledge, qualifications and licensing to be legal....
You could always revert the Steam option back to the T/stat, if you still have it?

Mal.

[KJM]

Sorry to hear about the premature death. The units Suyi ships are made in China like the cheap and cheerful jobs you get on ebay, but they aren't the same. I have a bunch of PID controllers, and the Auber ones are on par with Fuji PXR-3's. The cheap ones are most definiretely not.

Given the rattle and the smell - I'm guessing the switch mode power supply has let go. I've not had a Fuji or Auber PID die that way, but that is how the cheap ones go. This usually results in a lot of volts going into the board, and it becomes junk (junkier than it was ).

You won't have another drop-in 1/32nd DIN unit work the same as the Auber one either. There are some 'tuning' constants in the PID algorithm, and Suyi spent some time getting them right. If you put a generic PID in there, it'll need tuning. They usually have an auto-tune mechanism which will get you close to a good set of values, BUT - (there is always a but!) the approach used typically is a variation of the Ziegler-Nichols approach and this causes the controller to oscillate about the set point for a while. This is NOT a good thing for a pressure vessel.

You could auto-tune to a lower set point - which would be safe - then wind the temperature up to steam, but that won't give you good control at all. The thermostat might be better. The energy required for the phase change of water --> steam won't be taken into consideration.

Even if you had the old PID in working condition and went through and recorded all the data you can tune, I think the inner control loop has been tweaked - so the generic controller Auber lists isn't the same. I know the early one I had in my Silvia was definitely NOT the same as my generic Aubers - the constants were waaay different - if you put those constants into the generic unit it didn't do the same thing

Best of luck. I'd be putting the thermostat back for the moment....

/Kevin

[Dimal]

Auber ones are on par with Fuji PXR-3's

Sorry mate, can't agree with that. No Auber unit I've disassembled can compare with Fuji units I used years ago. Can only imagine that Fuji units are even better there days...

Mal.

[KJM]

Sorry mate, can't agree with that. No Auber unit I've disassembled can compare with Fuji units I used years ago. Can only imagine that Fuji units are even better there days...

Mal.

Sadly not so. Once. But certainly not now.. They're made in (I think) the same place Visually identical, but components might be sourced differently. I suspect not, but it may be so. The supply chain in the country they're made in is very, very dodgey. The QA process is probably quite good though, for these units. Manufacturing costs have been driven south by using every trick in the engineering book but also by offshore manufacturing. A colleague has a manufacturing site in that country and supply chain has been their biggest bugbear. They happen to be in the same factory complex as the manufacturer of the two devices in question

Anyway - back to the subject at hand - the quickest get-up-and-go is definitely to put back the steam thermostat (or thermostates..) and run it from there. Getting anything in during the silly season will be a pain.

Hope you get the machine up quickly again - the family christmas event sure ran my Expobar flat chat yesterday... The extended coffee snob family would riot if I didn't make them coffees!

/Kevin

[Dimal]

Hmmm...

Sad to hear that with the Fuji hardware...
They were among the most reliable 'stand-alone' units we used (in a process/SCADA installation that totalled thousands of loops) years ago....

How about the Watlow units mate? Does your colleague have any insights on those, or Eurotherm perhaps...

I'm going to be throwing the controller fitted to my DJE DB into the bin next year and was thinking of using either Fuji or another decent unit in its place.

Mal.

[Korbah]

Thanks for all the advice guys.

Seems I'll be forking out for a new Auber - frustrating given the initial $250 13mths ago.

Oh well

[Dimal]

Have you tried contacting Auber yet?

I think you have a valid argument for a replacement given the early failure of the original.

Mal.

[KJM]

I would be speaking to Suyi - he is a very reasonable person. Send an email and point out what happened. I had a DOA controller from him once that got instantly replaced $250 seems a lot more than a PID unit ought to cost... I actually have a unit in my junk box, but it has one of my cryptic markings on it - a "!" which usually means something is not working right and is a mental note for me to investigate. Unfortunately that was a long while ago, so I'm stuffed if I know what is wrong with the thing... otherwise I'd send it to you..

@Mal - no data on Watlow's. I actually have one, but I've never taken it apart, as it is in continuous use Which is slightly unusual for me But the fact that most of the competition is now manufactured over 'there' probably means they are too... Exactly what happens when EVERYTHING is made in the cheapest manufacturing place on the planet and no-one remembers how to make things is a problematic point to ponder...

/Kevin

[Dimal]

Yes mate, mirrors my concerns to a tee...

Hmmm, looks as though when the time comes, it's going to be a cross my fingers exercise.
Oh well, them's the breaks...

Mal.

[Korbah]

I haven't contactd Auber yet given the holidays etc. Their website does state that they require the PID to be returned to them for in-warranty repairs which will cost a decent amount to ship from australia to usa and take a fair while - it might be better to just pony up the cash and have a working coffee machine within the week. Bad timing with the holidays

The $250 was for the whole pid kit - it's ~$90 + shipping for the PID alone.

Thx for all the advice though everyone. Has stopped me getting the chinese models.

[KJM]

Just to finish - I checked out the PID I have. It isn't what is on the box It is unprogrammed! This was a unit I got some final year engineering students to play about with, and it needs the PID code put back into it... Need to find my programming cable for that... So not much use to you, I'm afraid.. I've made a mental note to try to get better at writing a sensible description of stuff in the junk box.

The generic Chinese models you refer to will work, but not like the one you're replacing. I'm not entirely sure I made that clear enough... They will control a resistive load to a set point, but they won't work as per the Auber unit. So you'd need to futz about a bit to get the performance to what you had. It can be done, but would involve a fair bit of playing about. Most of these units permit you to play with the constants in the PID control loop, and have other magic controls their designers threw in for good measure. You can look at the on-line information about what those constants do and how you tweak them, but the actual code inside the microcontroller is likely to be far less simple than that.. Controlling the temperature of the boiler is something that the generic PID controllers will have a small issue with because they tend to be a bit overly simplistic.. There is oodles of control authority in a Silvia for moving temperature in the boiler UP, but there is very little for going DOWN - you rely on heat loss. That kind of asymmetry in the way it works can cause poor performance from excessively student-project-like PID controllers

Best of luck with the playing about. I remember my PID'd Silvia fondly. A very robust piece of work. My Expobar is similarly 'brick sh*thouse' in robustness

/Kevin

[Javaphile]

How about the Watlow units mate?

My (~15 year old) Primo roaster has a Watlow in it which is the original and is still running perfectly.


Java "RED!" phile

[Dimal]

Good to know JP...

I wonder if the new ones are still as reliable...

Mal.

[Korbah]

so an update:

Contacted Auber who have shipped me a replacement unit at a very reasonable price. Just received it and plugged it in - espresso bliss

I'll test the steam function tomorrow but early signs are good! Plugged it in exactly like the old unit and it turned on immediately (vs. lying there dead and sad)

Thanks for the advice here guys and thanks to Suyi from Auber for the excellent service.