Results 1 to 19 of 19
Like Tree4Likes
  • 1 Post By MrJack
  • 3 Post By TOK

Thread: Troubleshooting Centre Channeling

  1. #1
    Junior Member Raraka's Avatar
    Join Date
    Oct 2012
    Posts
    6

    Question Troubleshooting Centre Channeling

    Gene Cafe Coffee Roaster $850 - Free Beans Free Freight
    Hi all,

    Been having some trouble with my pours, and was hoping somebody here might be able to shed some light on the situation.

    I'm using a ~12 month old Saeco Via Venezia, which is a single-boiler 15-bar machine with no 3-way solenoid, priced around the $450 mark. It came with a pressurised portafilter, but I've upgraded to a non-pressurised one from SeattleCoffeeGear. It's one of the older models, before they fitted the panarello steam wand, so it can actually produce decent milk. Admittedly it's not the greatest machine in the world, but I'm living within my means! I was quite content to stick with my aeropress until I graduated and could afford something decent, but I scored the Via Venezia brand new on eBay for $35, from a lady who won it in a raffle but didn't like coffee. So can't complain, really!

    The grinder is a 2nd-hand doserless Sunbeam Cafe Series EM0480 burr grinder.

    I'm finding that when I pull a shot, I'm getting a strange sort of "donut-shaped" impression in the centre of my puck, pictured below. When I tap the puck out, it tends to separate into two layers, as if it's beginning to channel, then spreading out and "splitting" the puck with a plane of water half-way down. There are no obvious soggy or dry spots, however. I don't think the impression is due to contact with the shower screen either, as the shape doesn't match up.

    In terms of my dosing and distribution method, I measure a shot glass full of coffee (Which is about 20g with these beans) grind them into an old coffee plunger, give it a twirl to break up any clumps, then tip that into the filter basket. I give the side a few taps with the tamper until it looks even, tamp with an ever-so-slightly-convex tamper using 30lb of pressure, and brush off any excess.

    I find my machine runs a little too hot, so I usually flush it until the heater light turns off, wait for it to come back on, then give it about 20 seconds more before I pull the shot.

    The extraction time is pretty consistent; the shot from the puck pictured below got me about 40mL in 27sec, with 20g of coffee, although it started to blond early at about 17 - 20sec. Flavour-wise, although it's got a slightly thin mouthfeel, there's definitely a good depth of character, but it's hidden underneath a heavy dose of sourness.

    I've tried using the Weiss Distribution Method as a solution, but it's made absolutely no difference. I've ordered a bottomless portafilter from SCG to try and get some more info, but at the moment, I'm stumped! Any thoughts would be greatly appreciated!


  2. #2
    TOK
    TOK is offline
    Senior Member
    Join Date
    Apr 2013
    Posts
    697
    Hi there.

    My opinion:

    You are attempting to use a domestic boiler machine as if it were a semi commercial or commercial type (say HX) machine.

    Firstly I think you are grinding too fine and tamping too hard, for this machine. Machines of this type like a grind that is a little more coarse and a less heavy tamp, than semi or fully commercial machines. And the lack of group valve on this to exhaust the pressure in the group handle at the end of the shot, results in what you see in the spent puck. That's just the way it is....

    Guessing that the "donut" (your words) in the puck is the continuing push of brewing water from the distibution nut after you stop the shot (due to lack of group valve to exhaust pressure.

    Also reading your modus operandi, it seems that while you think you are flushing out the super heated water before brewing the shot, you are actually allowing it top build up again (the electric element is quite powerful given the very small quantity of water it is heating in the tiny boiler). Think about it. You flush super heated water out. The light goes ON (meaning it is heating up again). You leave it for 20 seconds and brew....by which stage the water is way up there again.

    When I used these machines (and when I was starting out in coffee machine sales these type of machines were all there was unless you wanted to use a stove top), I just flushed the super heated water out until it stopped bubbling / ran quieter, and then made the coffee straight up...light on or not.

    Hope that helps, and am interested to hear what others may advise.

  3. #3
    Senior Member Journeyman's Avatar
    Join Date
    May 2009
    Location
    Bendigo
    Posts
    1,028
    The em0480 will grind direct to the pf but you might want to remove the pf holder.

    Try doing progressive tamping. It evens out the profile through the puck. And it's eaier to tamp with accuracy when doing it at ~3kg instead of 15kg.

  4. #4
    Junior Member Raraka's Avatar
    Join Date
    Oct 2012
    Posts
    6
    Thanks for the advice TOK. I'm confused though - You say I should grind coarser -and- tamp with less pressure? If I do that, the water will rush through the portafilter like the Niagra Falls, and I'll get a cup of pale brown water. My understanding is that, given I'm getting about 40mL of espresso from 20g of coffee in 27 seconds, my grind and tamp must be about right?

    You could be right about the impression in the puck being due to residual pressure after the pump switches off, but I'm not sure! It never used to do it with the old pressurised portafilter, although to be fair it was hard to tell since the "puck" I got from it was usually just a soggy mess. I wonder if there's anyone else with a similar machine who gets the same phenomenon?

    The light on my machine turns OFF when the heater is on, by the way. When the light is ON, it means the heater has turned off and the machine is ready to use. The rationale behind the way I flush the boiler is that the heating element is either running at 100% or is completely off. There's no in between/keep warm setting. So when the light switches on, it means the water is above a maximum threshold temperature, and power to the heating element is cut until the water cools to below a minimum threshold temperature, where it switches back on.

    In the case of my machine, that maximum threshold is slightly too hot for making drinkable coffee, so I always wait at least 20 seconds after the light comes on before I switch on the pump, to let the water cool down a little. This happens quite rapidly, since the boiler in this machine is -tiny-.
    By the time I am getting ready to pull my shot, the light will have already been on for a while (Meaning the water is probably too cool), so I "reset" the machine to a "known" point by flushing some water through until the light turns off, waiting until it comes back on, then waiting a further 20 - 30 seconds. I've done a lot of experimenting to achieve the right temperature with this method, and have been getting good results with that wait time.

    Journeyman: I've not tried progressive tamping, but I'll read up and give it a go. What are your thoughts on that method vs. WDT? Accurate tamping is probably easier, although I think my technique is already reasonably good.

  5. #5
    Senior Member Barry O'Speedwagon's Avatar
    Join Date
    Dec 2011
    Location
    PRL
    Posts
    2,585
    Is there scope for you to increase your dose Raraka?

    And FWIW the so-called 'standard' brew ratio is approx 40g of liquid espresso to 20g of ground coffee. This will equate to more than 40mls of liquid due to crema.......but the ratio that you describe is well within the bounds of what many people's tastes may enjoy, and probably closer to what I tend to make.

  6. #6
    Senior Member Journeyman's Avatar
    Join Date
    May 2009
    Location
    Bendigo
    Posts
    1,028
    Weis or NESW or other versions of tamping around the edges are to compensate for grinds dosing mainly to the centre of the puck. This means usually you are getting edge channelling, so compressing the edges (nutating tamp is another) means less of that. This doesn't seem to be your issue.

    I think you have 2 problems. 1 is the centre cavity you are getting. 1st thing I would check is your screen above the basket - is it open all the way across the screen or might some of the holes be blocked? It does look like you are getting pressure jets into the centre of the puck at some point but to get what you have in the photo I'd be more inclined to think it is at the start, not the end.

    Logic is, at the start you have a dry puck compressed at maybe 15kg. At 9-bar the water hits it HARD, much more than what you do with the tamper. From what I have learned the tamper is NOT about compressing the puck, it is about removing the air gaps and making the puck into a (relatively) homogenous whole that the water then enters under pressure. There's a couple of videos on here with glass baskets/pf's where you can see the action. It seems clear from them that almost all the activity happens at the start of the pour. Once the water pressure hits, that is a solid block you have in the basket until it releases.

    So if you are getting a cavity in the centre, to me that suggests there is water hitting FIRST in the centre.

    Then you have a different problem where your puck is breaking in half when you knock it. I think you are right about your 'level' idea - the top of the puck is being compressed and the bottom half? third? is not. When you knock it the compressed block falls out.

    Is your basket one with a shelf around the side? i.e. 58mm down to about half way then a ridge and it slopes in to maybe 54mm at the bottom? To me that would give the results you see. When you tamp, the shelf stops the pressure being transmitted into the bottom section so you have 2 different pressure profiles. When you knock it, the hard puck falls away.

    I use progressive tamping always, and for the last tamp I use a nutating action at minor pressure (as in less than the 3kg I use for the earlier ones) then polish. It seems to me the edge tamp methods ONLY resolve edge channel issues, and they cannot work to fix the pressure differential between top of puck and bottom of puck. So I progressive tamp to get consistent pressure through the puck then nutate to fix any edge issues.

  7. #7
    Senior Member
    Join Date
    Dec 2010
    Posts
    395
    what does the water stream look like if you run it without a PF? have you taken the shower screen and jet breaker out and cleaned them?

    also, I don't think that the water hits it "hard". the initial water pressure, and water velocity, is not very much.

  8. #8
    TOK
    TOK is offline
    Senior Member
    Join Date
    Apr 2013
    Posts
    697
    "...Thanks for the advice TOK. I'm confused though - You say I should grind coarser -and- tamp with less pressure? If I do that, the water will rush through the portafilter like the Niagra Falls, and I'll get a cup of pale brown water. My understanding is that, given I'm getting about 40mL of espresso from 20g of coffee in 27 seconds, my grind and tamp must be about right?..."

    This isn't a commercial machine...the "spec" is for commercial machines or machines with some "commercial type design features eg something having a group electro valve and different internal design specs. You will get the type of coffee or "behaviour" that a machine like yours can produce, but not like a quasi or full commercial type coffee.

    ".....You could be right about the impression in the puck being due to residual pressure after the pump switches off, but I'm not sure! It never used to do it with the old pressurised portafilter, although to be fair it was hard to tell since the "puck" I got from it was usually just a soggy mess. I wonder if there's anyone else with a similar machine who gets the same phenomenon?..."

    The pressurised group handle is very forgiving and made for beginners really. It is not intended to have fine commercial grind coffee packed hard in it. If you use it as it is intended by the manufacturer (not commercial fine grind and not pack tight), it easily allows the escape of pressure through the grinds when the group switch is deactivated, and when used like this (as intended) it always results in a sloppy puck. Normal.

    Its been a long time but from memory, the very simple arrangement in the group of these is the centre nut is the water distributor (so not just a lock nut for the shower screen) ergo.... that central position is where, if the machine is pushed to its limits (as it appears to be here with finer grind and harder tamp than is intended with such a model), you will get a direct indication of water under pressure coming into the puck and probably not distributing outwards evenly because the grinds are packed up too tight. It doesn't have the design feature to distribute the brew water as evenly through the grinds as other machines that cost more money.

    "...The light on my machine turns OFF when the heater is on, by the way. When the light is ON, it means the heater has turned off and the machine is ready to use. The rationale behind the way I flush the boiler is that the heating element is either running at 100% or is completely off. There's no in between/keep warm setting. So when the light switches on, it means the water is above a maximum threshold temperature, and power to the heating element is cut until the water cools to below a minimum threshold temperature, where it switches back on...."

    Again....its been a long time, but it sounds like something is awry? When you switch on a cold machine, in addition to some kind of "on" light switching on and staying on all the time the machine is on, the element light should switch on to indicate....that the element is switched on! The element light should go out when the element has been switched off by the thermostat, and it will go on and off as the machine cycles normally. Ergo my explanation above. And perhaps I've got it wrong and the light on the Via Venezia may be a "ready light" instead, as in the machine has gone to temperature, the element is switched off, and its ready to go....?

    Regardless...This is a basic but sound and reliable beginner machine. If you expect to do things, and achieve results, per commercial type spec, you may need to consider an upgrade to something more suitable.

    Honestly I would consider this:
    a) I have no idea how you can cram 20 grams of coffee into that filter....far too much.
    b) Try coarsening the grind a tad with an appropriate, smaller dose. I havent weighed what such a filter can take, but I am guessing it wouldn't be more than 16 grams....in any case I do not weight my doses.......suffice to say that for any given grind setting, after tamping you should find the tamped, levelled dose in the puck to be say approximately 4 mm below the top edge of the filter (given small differences between models). Therefore the weight of the dose will vary with the grind setting, to achieve the same height in the filter after tamping.


    Hope that helps.
    Last edited by TOK; 28th April 2014 at 10:32 PM.

  9. #9
    Senior Member Journeyman's Avatar
    Join Date
    May 2009
    Location
    Bendigo
    Posts
    1,028
    @Hildy - For conversions, 1 bar = 14.5 lb/in^2, so 9 bar for a coffee machine is 130.3 lbs/in^2

    Given a 58mm basket has about 5 in^2 area, that's around 678 lbs/in^2 hitting that puck. The 30 or so lbs from a tamp is barely a blip on such a scale.

    Now you might be saying that this particular machine brings up the pressure very slowly but that would only be comparative. I'd think the jet of water would be travelling at significant speed when released from the boiler, because to achieve 9 bar at the puck, the boiler has to be pushing the water through at at least that pressure. i.e. I think the water would definitely 'hit' the puck by any normal definition.

  10. #10
    Senior Member
    Join Date
    Oct 2011
    Posts
    1,376
    Given an approximate flowrate of 30mL in 30s, the velocity in even the 6-9mm tubing to the group is tiny (< 0.05 m/s by my back of the envelope calc). So no, it won't "hit hard", in my opinion.
    chokkidog likes this.

  11. #11
    Senior Member
    Join Date
    Dec 2010
    Posts
    395
    velocity in the pipe should be about 1-10 cm/s (depending on the diameter of the pipe) to deliver 1 ml/s.

    not just that, the jet breaker and shower screen disperse the water so that it's more like a sprinkler than a water jet.

  12. #12
    Senior Member Journeyman's Avatar
    Join Date
    May 2009
    Location
    Bendigo
    Posts
    1,028
    @MrJack - you may have missed Talk_Coffee's video at http://coffeesnobs.com.au/brewing-eq...rtafilter.html

    Note the sudden 'shock' to the grinds as the water starts? Keep in mind the video seems to be slow motion if the water movement is any guide. That seems a rather abrupt change due to a rapid increase in pressure.

    Even so, that isn't the issue for the OP. I'm just pointing out what seemed to be an erroneous assumption in a post, but the real issue is the hole in the centre of the puck. Given the puck is pressurised and soaked by the 9 bar pump, it seems logical to me that the damage is being done at the start, when the puck is pressurised at a measly ~15kg and that damage has to be caused by the water preferentially hitting the centre of the puck.

    That was why i questioned at the start about whether the screen was open out to the edges - it seems clear that the water is hitting that centre section under pressure BEFORE there is a layer of water to spread out the effect. i.e. the puck is being hit hard by the water.

    I'd be curious as to how you calc'd the <0.5m/s - it seems to me the entire space above the puck plus more is being filled in much less than 30secs because on most machines the water not only fills that space, but also the grind volume within maybe 4 or 5 secs - we know that because that's about when the coffee starts flowing out of the basket.

  13. #13
    Senior Member chokkidog's Avatar
    Join Date
    Apr 2008
    Location
    Victoria, Australia
    Posts
    2,095
    All due respect but that doesn't look like a 'hard hit' to me.... Low dose, little if any tamp, yet the water fails to blow the coffee apart,
    it seems to wash/float across the top of the dose and up the sides of the pf.

    Most vibe pumps take up 6-8 seconds to come up to pressure once the pump is activated, a 'hard hit' would see the pressure gauge slam
    to 9 bar? My rotary pump takes about 3 seconds but doesn't cause any deformities to the puck.

    Also getting back to the OP.. if the bottom of the puck was wetter than the top then the weight of water and grounds would cause the whole
    puck to fall out?? It looks to me as though the top of the puck is wet and the bottom dry?

    Speculation, of course. :-D

  14. #14
    Senior Member Journeyman's Avatar
    Join Date
    May 2009
    Location
    Bendigo
    Posts
    1,028
    The coffee gets 'hit' with the pressure before the water gets there. Unfortunately it is difficult to say much else with regard to the OP because the system in use seems to be a manual lever - thus the lift of the grounds followed by the compression as the pressure comes on. I wouldn't expect a semi-auto system to have the lift, and I'd also expect the pressure to hit before the water. So to excavate the centre like that, it seems fairly obvious the water has to be squirting onto it from the centre of the screen above.

    My pump doesn't deform the puck either and I have the 1st drips coming out of the naked PF in about 5 secs. I don't have channelling and it's been a lng time since I had anything but a single stream after the 1st couple of seconds of drips.

    Near as I can see, nobody else has come up with anything else as a cause for that puck centre except for the screen issue, which is why I mentioned the initial hit compressing the grinds at far greater than mere tamper pressure - that pressure means, as near as I can tell, the cavity is formed BEFORE the coffee flows, not by a failure to release pressure at the end.

    TOK mentions it might have the water coming through the centre nut, but then you'd expect that anything EXCEPT a hard tamp would get a central cavity, rather than it being caused by a hard tamp as s/he suggests. Also the central hole is definitely a donut shape so the nut would have to be more like a hose sprinkler than a nut with a central hole.

    Grain physics suggests the ability of grains to move in a fluid (air or otherwise) needs an initial expansion to 'unlock' the grains from each other - it seems difficult to see how this would happen with an increase in pressure as the water comes in. I'd think that would lock the grains in place more.

    The split puck is more complex to understand I think, but with a concentrated 'jet' scooping out the hollow at the start, I'd expect the water to channel through the centre first. While the coffee is pouring the upper layer would be soaking up water. Then at the end you'd have a centre that provided most of the shot, a wet top half (or thereabouts) and fairly dry sides. I don't know what the basket is like (i.e. does it have a ridge? The pics don't show enough to tell) but it's not difficult to envisage water finding its way down the path of least resistance and causing a funnel effect of water towards the central channelling bringing 2 different 'textures' to the puck.

    The picture seems to support such an idea - the part that drops out has a convex bottom to it and you can see the channel in the part still in the basket.

  15. #15
    TOK
    TOK is offline
    Senior Member
    Join Date
    Apr 2013
    Posts
    697
    You blokes are a riot !

    Suggest we all forget the technical mumbo jumbo including my own, and let's take the OP back to basic principles. This is a very simple machine.

    a) suggest you remove the group shower and makes sure it is all clear throughout ie when it is held it up to the light, you can see light shining through the holes all round. If not, (ie if it appears to be blocked in some way) place the screen on a cook top and wait for the carbon/coffee build up to burn out. You will get a *green* flame first up (if on gas) that will settle to a regular flame. It can be done on a normal electric ring. After a few minutes pick the screen up off the cook top with a pair of plyers, and wash under a tap. It should now be completely clear when held up to the light.

    Refit the shower to the machine, first making sure the area of the group over which the shower screen fits, is clean and not gunked up with rubbery coffee dregs (ie clean the area before refitting the shower). Same deal for the screen lock nut…make sure it is clear and clean.

    b) Next / work out the correct VOLUMETRIC DOSE for the filter. For any given grind setting, grind dose and tamp the puck into the group handle. The level, tamped dose should be roughly say 4 mm below the top rim of the filter. DRY THE GROUP. This is important, so coffee grinds don’t stick to the shower when you fit up the group handle, and ruin the following experiment…
    Fit the "loaded" group handle to the group. Remove the group handle from the group and see if there is a shower imprint on top of the puck. If not, it means you need to add a touch more grinds. If there is a ”whisker” of an imprint, your doe is correct, and if the the group handle was hard to fit up Or if there is a heavy imprint, then the dose was too much (too high).

    Using this method, for the particular grind setting, work out what is the correct dosed height in the filter. (the dose where there was just a whisker of an imprint).

    Re apply the group handle to the group and make coffee. Decide whether the grind is too fine, too coarse or correct. If you decide to make a grind adjustment, note that going finer (for the same weight of beans) will settle the fines more and reduce the volumetric dose in the group handle. If you want this to work properly, always work to the same dosed height in the filter.

    So in essence, in order to work out the right grind for any particular coffee, you need to grind dose and tamp first, always dosing to the right height in the filter basket, make coffee, and work back to the grind to make adjustments.

    Having decided on what the correct dose looks like in your group handle, then make coffee, varying the grind and tamp (but always working to the same volumetric dose) to find what works best for your machine.

    The pressure of the tamp is not necessarily important, but that you tamp to a consistent pressure each time you make coffee….is important, although as stated in a previous post I don’t think that tamping to commercial machine type spec (30lbs) is good with your model.

    In essence, re-learn to use your machine, now that you have changed the group handle to a different type than was supplied originally.

    I venture a guess that once this has been worked out for best effect with the machine, there will no longer be any “donutting” (again…your term) in the puck, but it will never look like a "regular" puck from a machine that has a group valve, because it doesn't have one...

    And going back to something I said in a previous post, you cant expect to get commercial type coffee from this machine, but you can get the best that you can from it…


    Hope the helps.

  16. #16
    Senior Member
    Join Date
    Dec 2010
    Posts
    395
    Quote Originally Posted by Journeyman View Post
    @MrJack - you may have missed Talk_Coffee's video at http://coffeesnobs.com.au/brewing-eq...rtafilter.html
    Note the sudden 'shock' to the grinds as the water starts? Keep in mind the video seems to be slow motion if the water movement is any guide. That seems a rather abrupt change due to a rapid increase in pressure.
    The pressure is close to zero at that point because the system hasn't had a chance to pressurise.

    Even so, that isn't the issue for the OP. I'm just pointing out what seemed to be an erroneous assumption in a post, but the real issue is the hole in the centre of the puck. Given the puck is pressurised and soaked by the 9 bar pump, it seems logical to me that the damage is being done at the start, when the puck is pressurised at a measly ~15kg and that damage has to be caused by the water preferentially hitting the centre of the puck.
    That was why i questioned at the start about whether the screen was open out to the edges - it seems clear that the water is hitting that centre section under pressure BEFORE there is a layer of water to spread out the effect. i.e. the puck is being hit hard by the water.
    Via venezias don't actually have any water hitting the centre of the puck because they have a screen screw (it's a little phillips head thingy, not like the silvia's hex head bolt).

    I'd be curious as to how you calc'd the <0.5m/s - it seems to me the entire space above the puck plus more is being filled in much less than 30secs because on most machines the water not only fills that space, but also the grind volume within maybe 4 or 5 secs - we know that because that's about when the coffee starts flowing out of the basket.
    The space between the shower screen and the basket is about 5ml. I haven't weighed a wetted puck to see how much air gap there is in the actual puck.

    the velocity out of the shower screen can't be more than a few mm/s at the start because of flow-pressure curves on the ulka pump (it can produce about what, 6-7ml/s against zero head, or a shower screen velocity (ignoring inertial and viscous losses) of 1-1.3cm/s. The droplets shouldn't be able to accelerate with gravity very much in the headspace.

    I have no idea how things are 'hit with pressure' like you're saying, though.

    I don't know enough about grain physics to comment on puckology, but I do know a little bit about hydrodynamics.

  17. #17
    Junior Member lucasr53's Avatar
    Join Date
    Mar 2014
    Posts
    14
    I'm definitely with journeyman as far as the donut goes.
    Check your shower screen. Give it a good clean, and a wipe over between shots.
    Maintenance n cleanliness for me are always my first points of call with shot issues like this. It's pointless getting into complex details and techniques if your fundamentals are not right.
    As far as the nut goes, the only way you'd get that result is if it wasn't screwed tight so the water was just pouring 'through' the nut and not the screen.

    So, check the screens working properly first. Let us know n then we can get into more details about techniques and tech stuff. I tend to find its the simple solutions that get overlooked.

  18. #18
    Senior Member
    Join Date
    Dec 2010
    Posts
    395
    I can't remember what the saeco group looks like anymore, but if it's so loose that you get nut screw water, the entire shower screen should be falling out. (with the design from the silvia).

    I should actually take mine off the shelf, refurb it, and give it away.

  19. #19
    Senior Member
    Join Date
    Oct 2011
    Posts
    1,376
    Behmor Brazen - $249 - Free Freight
    Quote Originally Posted by Journeyman View Post
    The coffee gets 'hit' with the pressure before the water gets there.
    I'm curious to know how you came to that conclusion? The air pressure will only rise as fast as it is compressed by the water (and some of it is probably going to be expelled through the puck). Plus, the pump discharge pressure doesn't start at 9 barg. It will rise, following the system curve, as the downstream flow restriction increases.

    It's a bit beside the point anyway; has it even been established that the puck deformation doesn't occur when the PF is removed? Perhaps the centre channeling is unrelated?



Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •