Seasoning a coffee machine

By seasoning a coffee machine – I am talking about pulling a few shots per group having done a chemical clean. Conventional wisdom says that if this is not done then the coffee will have an unpleasant metallic tang to it, because the machine is “too clean”.

Something about this just doesn’t sit right for me. I am going to assume that if you are reading this then you probably clean a coffee machine the same way that I do: portafilters regularly scrubbed and immaculate, screens dropped and scrubbed clean, dispersion screw kept clean, group seal cleaned and dispersion block scrubbed clean. After all this we do run the backflush routine with a small amount of chemical.

For years I did the backflush bit wrong.  1  These days I find the routine on the side of a bottle of Full Circle the best explanation and method.  2

Now as I understand it we are really trying to clean a small section of tube that runs between the solenoid valve and the dispersion screw.  3  This tube gets dirty because coffee travels up it during the pressure release at the end of pulling a shot when the solenoid opens up.  As such this tube can get a build up of unpleasant flavours.  This happens very slowly – compared to how quickly the dirt in a portafilter builds up and ruins a shot.  Flushing in between shots probably helps slow down the build up in the tube.

backflush

So when we backflush we switch the group on two pump in water to dissolve some chemical and then switch the group off, the solenoid pressure release sucking the chemical up the tube where it can start to strip off any build up.  We repeat this process to ensure thorough cleaning and then we go through a rinsing process.

I think we all agree that at this stage there is no chemical left behind, and if there were it would be a specific and unpleasant taint to the next shots.  However we are told that the next shot will taste bad regardless.  I would like to suggest that this bad taste is primarily due to a loss of heat at the group, affecting the brew temperature and creating an unpleasant underextraction.

First of all – this piece of information is pretty much written in stone but how many of you have tasted the next shot out?  Exactly how did it taste bad?  Very few people have tasted it because they are always told it is going to be bad, so why would you?.  Those of us that have have probably done so in one of two scenarios:

1).  Tasting immediately after a backflush.  Here the group has likely cooled down, because very little water has moved from the boiler out of the front of the group.  On some machines this may influence the brew temperature.   Plus we’ve left our portafilters out for a while and I would contend that the first shot you brew will be under temp.  We may have gotten into the habit of using that shot to bring the group and portafilters up to temperature, because we know it is a wasted shot regardless.  4

2).  The next morning.  Here the group has again cooled down due to a long period without water travelling from the boiler to the dispersion screen, and the shot would likely be undertemp.  If you leave a stock Linea alone for 30 minutes during the day you need to flush a lot of water through it to get it back up to maximum, stable brew temp.  What if you left it 8 hours?

In both cases it is somewhat likely that the poor taste of the shots could be primarily attributed to brew temperature.  I asked Josh Dick from Urnex this very question during our Barista Magazine article and there didn’t seem a more likely explanation.

I have a Synesso Hydra.  From leaving the pump to hitting the coffee the water travels through a good deal of metal piping.  Are we seriously suggesting that while it will pick up no metallic taint from the majority of its journey, but if a small tube near the coffee is not coated in coffee oil that it will somehow have a massive impact on taste when the rest of the machine’s metallic surface will not?  If that tube is dirty then I absolutely believe that it will have a negative impact on the shot, but too clean?

Last night I did a very simple experiment.  I chemically cleaned my machine, and left chemical in it for well over 2 hours.  I want it as clean as was possible.  This morning I dialled in the grinder on the left group, and once happy I then flushed the right hand group up as hot as I could.  Once I was satisfied that it was hot enough I pulled a shot and I drank it.  The seasoning shot.  Did it taste bad? No. Did it taste sour or metallic?  No. Could the fact that it was my first coffee of the day have influenced me? Absolutely.

However – lots of you reading this have coffee machines.  Lots of you probably have Scace devices.  I am absolutely happy to be proven wrong but I would just like to get to the bottom of this phenomenon.  How do seasoning shots taste to everybody?  Do we all agree on what is wrong?

  1. I would put chemical in, switch on the pump – assume that this part was the cleaning part and let is run for 30 seconds and then just rinse everything clean.  A complete waste of time.  ↩︎
  2. By the way – given the choice between Cafiza and Full Circle is anyone not opting for the sustainable option?  It may cost a little more but surely the tiny cost increase per backflush really doesn’t matter.  ↩︎
  3. If I am wrong here – I am more than happy to be corrected!  ↩︎
  4. I accept that on some machines backflushing and then immediate brewing may result in high/higher brew temperatures, but rarely are people pulling shots immediately afterwards – usually there is sufficient time for that extra heat to dissipate.  ↩︎

24 Comments

  1. Nice article. Just wondering why Full Circle is sustainable but Cafiza not? The ingredients for Full Circle are basically a mixture of sodium carbonates, detergent and a buffer, which is what I assumed Cafiza was, though they see fit not to publish the ingredients of Cafiza.

  2. Again you tackle the unquestionable and obvious James, very nice. While I am in the group of those that have blindly seasoned without question, I do it for a few different reasons than that of the machine being to clean. I do it because I am always wary that there may some last little bit of cafiza hiding up in the screen mainly and feel that the first puck takes care of that. Of course with that said, have I scientifically sat down and tested it? no… shame on me. I have however had an exagerated example of cafiza effect with the clovers several times. It is easy for folks to use to much cleaner in which case there is a need for multiple seasoning brews. We have a very regimented cleaning cycle for both espresso machine and clovers during the day, every hour for the Synesso and every 40 to 50 cups for the clovers. If we do not do a seasoning brew after cleaning of the clovers that chemically taste is very apparent. This is not a heat loss effect because the machine is about as hot as it gets when finished with a cleaning cycle due to pumping out multiple rinses. Another side note for the seasoning shot is that it has more meaning to folks who may have a less regimented cleaning cycle, shots may pull a little different (faster) after a clean depending on how dirty that screen was. I know I used to notice this in the early days before I realized how important cleaning was. I tend to use the seasoning shot the trinity of espresso quality insurance, a temp boost, a cafiza remover and a flow check.

  3. With Full Circle I think the manufacturing process to generate the ingredients is sustainable. They are the same ingredients I think, just produced in a more expensive/sustainable way.

  4. If the coffee tastes like cleaner then we should maybe ask if there is a better way of removing residue than using coffee. Maybe there isn’t, but perhaps worth looking at.

    Interestingly some people find that the first shot after a clean chokes a little/pours too slow. Improved flow rate does usually decrease extraction flow rate – something I’ve seen when flow restrictors have started to scale up on one group more than another. The decreased flow rate apparently compacts the puck less aggressively, migrating more fines and slowing the shot down. However there are many other factors that influence flow rate, and screen cleanliness is no doubt one of them.

    I should add that I am not saying we shouldn’t “season”, or pull test shots after cleaning. I am just suggesting that ‘seasoning’ maybe isn’t the best word to use (as many businesses may be aware of seasoning and believe this is a suitable use for older ground coffee!).

  5. I always thought it had more to do with the baskets and spouts (mainly the baskets) getting clogged a bit. Going to bottomless reinforced this for me as it stopped happening but again, there is a sad lack of actually testing these ideas.

  6. I arrive at the shop and the 1st shot out of the machine is my 1st coffee of the day (I developed this habit years ago before anyone ever mentioned seasoning to me). Assuming you get pour right first time I’m completely satisfied it’s as good as any other shot. On this basis we’ve never seasoned.

  7. One thing I might suggest you do:

    Take a dirty PF, one covered in coffee oils pretty badly, remove the basket, remove the spout and cover the hole so that you can pour the solution in the PF. Make some Cafizza/Urnex solution and pour in the PF, wait an hour or so. Now remove the solution and rinse with water, but DO NOT brush out anything. How much stuff is still left on the sides of the PF? I found that soaking doesn’t necessarily remove all the oils and coffee residue, so I’m not so sure backflushing with detergent leaves the pipes super clean.

    Oh you can do the above simply by soaking the PFs in some solution and checking how much soaking actually removes.

    Regards,
    Tom

  8. I always used to taste seasoning shots, mostly because I didn’t know why they always tasted so bad. Since I got back I’ve also started to think the ‘metallic’ taste that I got was likely due to below-temp extractions. I think saying that the brew path has to be seasoned or the coffee tastes metallic is along the same lines as saying spouts introduce a metallic taste that a naked portafilter doesn’t. The water touches too much metal on the way to the cup for this argument to hold much, er, water.
    Incidentally, what does this mean for my crepe pan seasoning routine?

  9. I’d agree group temp fluctuations (up or down) are probably the main cause for bad shots after cleaning; I used to have to make as many as 3-4 shots on our old machine before they started tasting about right, when if it had cooled off a lot after cleaning – from sitting idle for a while, or overnight – which I guess was the machine finding it’s overall temp balance, etc (most traditional machines make their best coffee during the busy periods, right?). I’m not sure about the metalic taste directly after cleaning, because, as I say, my 1st shots from cold or after cleaning hardly ever looked drinkable, so it’s something I have yet to experience properly, as I didn’t drink them (should have forced myself in order to test, but instead usually tried to grab shots when they looked at their best!). But if it does occur, the fact water goes past lots of metal before it gets to the coffee might not necessarily mean it isn’t a factor; perhaps it’s the hot, brewed espresso (rather than plain water) that can attract this metal-taste from the surfaces of the basket and p/f… I’ll look out for it.

  10. James I agree in that we are trying to clean the banjo tube on saturated groups and the path from dispersion to solenoid on HX machines. I have always been a screen and screw in backflusher as I believe agitation is vital to get things clean. Also I have been doing something similiar to the Urnex instructions (point 3) for a long time. If you think about it, if you just cycle the brew on and off numerous times without umlocking the pf, you are just forcing water up aginst a blind filter full of water. I think it is this unlocking and putting in an empty basket (and running water through) is very important.

    Having dissected quite a few groups it is also apparent that no amount of backflushing will ever get the pipe or dispersion block (brass) back to factory clean – unless stainless parts are used. Even if you dunk the entire tube or block in cleaning solution and an ultrasonic bath you still struggle to remove oils on the inside of pipes.

    Anyway back to the original topic. I would venture a guess that the seasoning shot was a product liability thing put in place by the cleaning companies for good measure, and also since we are dealing with foodstuff seasoning is a common thing.

    The crappy shot syndrome I would agree is to do with temperature of the handles more than anything. Personally I do a seasoning shot only because I want to use up the coffee that has been left up in the grinder chute.

  11. Played with this idea with a naked portafilter on a little home machine after a chemical clean. I’d rinsed the basket and the showerscreen as much as possible but there was a chemical taint in the middle palate and a sharpness in the aftertaste irregardless. I feel more than a little queasy now.. bleugh x_x

  12. Judging by the article and comments so far, it would appear we are saying the “bad first shot” phenomena has nothing to do with seasoning the basket with coffee? My limited experience indicates a super clean basket causes the shot to run fast and to “edge channel” as opposed to a basket that has only been wiped clean after a shot. Could this be the cause of the sub-optimal flavour of the first shot?

  13. James,
    I really appreciated the critical critique of conventional ‘knowledge’ there. I hadn’t thought about my cleaning process critically like that before. I have wondered though, does the time of day we clean our machines matter much? It seems conventional to purge the machine, etc. at the end of the day and run our first ‘useless’ shots of the day before opening. What if we cleaned the machine during the reasonably slower hours of the day around 3-4pm? Would this somehow change anything? You’ve created a monster of questions.

  14. looking back, i guess i assessed why i season and moved on without ever appreciating why, bad science!
    For me, it was always a way to ensure that the last of the cleaner has been removed from the group handle and head, making it more about poor cleaning technique (and training) than shot quality.

    certainly leaves a lot to ponder though.

  15. I rocked up to work the other morning and instead of the usual dumping of the first shot, I had a wee taste.
    The shot was reasonably palatable, and I must admit that this practice for us is often just to clear the grinder of beans that have sat in the grinder throat overnight. I still think I prefer the taste after a few shots, as the “clean” portafilters do lend a slight metallic taste, being well used exposed brass.
    A little of subject, but I’m wondering if your ebay Linea makeover from a while back was documented in a little more detail anywhere. I’ve just got an older 2 group AV and am having a few little issues, I would love to know what things you tweaked to get back in shape. Did your machine have any flow restrictors installed?, pretty sure this machine has nothing.
    Cheers Troy.

  16. I was at the London School of coffee on Friday when the subject of Seasoning came up. Morten described how the same charge that aided the extraction of the organic molecules from coffee could also lead to them being attracted to metals of the portafilter etc. As I say I only learned the basics of this on friday but it sounded very feasable and the diagrams were great! Maybe Morten could shed some more light?

  17. Not sure about your fancy commercial machines…but on a home machine, I always thought that brass was a horrible smelling/tasting metal especially when you scrubbed down a portafilter free of its chrome plating. Spritz in a little hydrogen peroxide to enhance the flavour molecules, Yuck!
    Brass! -horrible stuff!
    Best to build up a patina with a few shots or so…
    On a home machine..coffee residues build up and get baked on as black tarry substances. They need to scrubbed out on a regular basis. Cleaning improves tastes greatly – just be careful of espresso touching bare brass.

  18. Very interesting challenge. My flow rate is actually better and faster after a good clean. I usually find myself having to adjust to a finer grind setting.

  19. hmm… interesting thread again James! I’m gonna try my first shot tomorrow morning, i would never think of this. What I usually do to season the machine is i put old coffee from yesterday’s grinder cleaning. When my grinder is clean I store all that coffee from inside grinder in a jar, so basically I’m using coffee that would go to waste…

  20. mmm…you raised the issue of seasoning a coffee machine so as to not taste metallic…..however then you talked about how the espresso tastes after back flushing. In the beginning of the article you mentioned chemical cleaning after pulling shots in the process of seasoning a machine…why?? Aren’t we trying to remove the metallic taste by pulling shots until achieved? isn’t tasting shots after back-flushing different and a separate issue to seasoning??

  21. The taste of metal may be a product of overzealous scrubbing with something abrasive. However, ample flushing would seem to resolve the issue. Personally, I think that too many baristas over think the coffee making process without tasting the coffee. The only thing that really matters is how the coffee comes across in terms of taste, texture, finish, etc.. Thanks, though, it’s good to bring these things up, burst some bubbles.

  22. Remarkable article. Cleaning the machines takes a careful process. It takes ample time to finish the job. However it is very important that the coffee is free of any taste of metal or lead. 

Submit a comment