It's a complete guess but my thought is that when ground and brewed, the lower atmospheric pressure lets more gasses escape and results in a sour shot. So with older beans there's less gas

What I know is that I was driving myself crazy with fresh beans. Then by accident I used some old beans and it got much better. Your mileage may vary, but it would be the first thing I suggest when brewing at altitude

I'm also sort of struggling with the Lindonium tool. I had some pretty great consistency (after a lot of practice!) using a straightened paperclip and no funnel. The Lindonium tool is bigger and less precise than a paperclip, obviously, so I'm having to adapt my technique. I was doing okay for a while being very gentle with my movements, to not spill out of the basket. Then I got a funnel (Tidaka 30mm) and so can be more aggressive with the way I stir during WDT. I've been really whipping it around, as if I were lathering up a shaving brush. This seems to product really bad results. Channeling is visually apparent from sprites, significant dead spots at the end of the pour, etc. The beginning of the pour typically looks good and starts evenly, which is puzzling. And I can tell the impact in the cup, with more bitterness and astringency.

I'm going to experiment with changing my WDT technique up pretty radically. I'm going to try stirring very gently, and see how that helps. I also like what was suggested upthread, to do small/local circular motions to break up the grounds with a final large/global circle to distribute evenly and level across the puck. If that doesn't help, I'm going to try eliminating WDT to see if anything improves/worsens.

Would appreciate any advice or insight folks have!

Perhaps follow what the manufacturer does? (You have to click on the graphic below to get to Reiss' Vimeo channel)

Thanks, that's a great video! Very interesting. That's pretty close to how I pulled my two shots since my prior post. Slower for sure, a little more local movement as opposed to global sweeping circles, though he does seem to do a little of that at the end. This seemed to help a little, but was inconsistent. I will have to continue to experiment with technique; before, without the collar, with very slow light movements, I got less balding

Before I was doing more of what we can sort of see rimblas do here, with the fast sweeping circles across the whole of the basket. I would do that a few times clockwise, spiraling in and out. It was with this method that I was having the worst dead spots under the basket.

Interesting from the Londinium video: he demonstrates that the tool makes the espresso flow slower. Maybe I'm confused, but isn't that the opposite of what we should be aiming for with our prep? If it's flowing slower, we need to grind coarser to hit the same brew ratio/time, and so are extracting less. If it flows faster, we can grind finder and extract more. Thoughts?

What he meant in the video was, I think, that with the same coffee, same grind, same dose, the first shot flowed way too fast, because distribution was uneven and so liquid flowed faster thru the paths of less resistance. That shot would taste noticeable different than the second one, where the flow was slower because it was going thru very close to the same resistance across the entire puck, getting a better extraction.

Once your distribution is good and consistent, then you can distinguish the different results of small changes in grind setting.

I think the tool itself works well because of two features: the triple wire & loop approach makes it much more effective and efficient (it would be a fair bit more work to get a single needle thru every portion of the coffee. The other, more subtle (and maybe a critical) feature is the flexibility of the wire. I can actually feel where the coffee is more dense as I WDT.

So I basically do what the video shows but give extra attention to areas that feel denser. I then raise the tool up so the loops are just in the top of the coffee, and brush around to get a roughly level surface. I also tap the PF on the tamping mat once, just before using my LevTamp.

I'll give it a try without the tap to see if that makes any appreciable difference one way or the other.

So I just tried to pull a couple shots again. The first one was with the Londinium tool and portafilter collar, trying to mimic the technique from the Lindonium video as much as possible. Again, I got extreme balding/dead spots about halfway through the shot, right as the flow rate starts to ramp up; and the stream broke up and skittered around the basket bottom. In the cup, I could taste bad channeling, a combination of sour astringency and bitterness.

The second shot I pulled without any WDT at all. I dumped the grounds in, leveled with my tool, and tamped. Overall it flowed slightly faster, but I saw the first droplets during preinfusion much earlier than with the Lindonium WDT. I still got some dead spots but less extreme than the first shot and occurring later; and the stream did not break up or skitter around. The taste was definitely better, this one was actually drinkable without significant bitterness.

Overall, not sure what to make of this. I'm sure there's still some inconsistency with my technique with the new tool, but it seems like I'm getting better results without it, or with my previous paperclip method. I'll keep working on it, I suppose. I think the best success I had was with Bret describes, stirring as in the Londinium video and then focusing on the top 1/3 of the puck at the end to produce a fairly level puck before leveling/tamping. I used to do a side shuffle sort of move to even it out, and then tapping on the counter to settle the puck before leveling. I have been trying to eliminate this to improve consistency, so maybe the lack of this has been producing worse results?

I definitely agree with you, Bret, that the design of the tool seems like it should work really well! I figure it must be operator error...

inexorable_end wrote:Overall, not sure what to make of this. I'm sure there's still some inconsistency with my technique with the new tool, but it seems like I'm getting better results without it, or with my previous paperclip method.

The most likely explanation for reduced flow with WDT is elimination of channeling. Any change in your espresso workflow will probably require re-dialing in your extraction. Since you're getting dead spots in your puck, you need to lower your dose and/or coarsen your grind.

Don't just pull one sink shot and leave it at that. Fix the mechanics of your extraction, play around with it for a bit, and then draw your conclusions.

I agree with John, I'd keep at it. It could be the machine, but I haven't had a single gusher since starting with my Londinium R and the WDT tool - and if I understand correctly, lever machine's water hits the puck with a lot more force (not sure if that would help or hurt channeling).

If you're getting tired of stirring, maybe try a blind shaker.

Thanks John, very helpful!

The most likely explanation for reduced flow with WDT is elimination of channeling. Any change in your espresso workflow will probably require re-dialing in your extraction. Since you're getting dead spots in your puck, you need to lower your dose and/or coarsen your grind.

That in particular helped me out conceptually. I was stuck thinking about grind quality, after experimenting with sous vide beans, where (I think) more unimodal grind -> less fines -> faster flow. But you're totally right, since with WDT we're not removing any fines, a better distribution of fines in the puck should lead to a more well-restricted flow as a result of less channeling.

I did try doing as you suggested, coarsening my grind. (I'm already at 16g in an 18g basket, so don't really want to lower dose further.) The first two shots I got were gushers, like 60g out in 35s including a 13s preinfusion. They didn't show bald spots, since the flow was heavy, but I did see significant and persistent sprites, which is another sign of channeling and one I haven't seen in a while.


So I tightened my grind back up to roughly where I began. I did WDT as I had been and got a solid shot with minimal/acceptable balding in the last ~3-5s and good taste. Not perfect, but definitely an improvement.

But -- I did change one other parameter with this shot (bad science, I know): my leveling tool was set too deep, since my prior bag of beans was denser, so I adjusted that to just barely level the surface. Since I didn't really change anything else significant with the puck prep process, I wonder if this was the source of my issues; now that I think about it, I switched over to this bag right as I got the collar that coincided the start of my issues. Maybe the leveling tool was too deep in the beans, and was dragging most of the puck around and half-tamping it, destroying my (optimistically good) distribution work from the WDT tool. I'll have to pull some more shots soon to validate.

Don't just pull one sink shot and leave it at that. Fix the mechanics of your extraction, play around with it for a bit, and then draw your conclusions.

For whatever reason I rarely dump my bad shots, I just choke them down. Probably because hand grinding is a pain. I'm definitely always thinking about technique and making small adjustments to process parameters, but iteration happens more slowly. So I took your advice here as well and dumped my two poor shots down the sink! The third one was the charm.

It could be the machine, but I haven't had a single gusher since starting with my Londinium R

Well, let's not rule this one out yet! I'm working with a Gaggia Classic with a meCoffee PID and a few other mods. Certainly serviceable, and I've pulled some damn great shots, but quite inconsistent and lacks real pressure profiling capabilities. I wouldn't be surprised if the machine was causing me a ton of issues I've been compensating for through puck prep, etc. Same with my Lido ET, which is adequate at best and which has a tenuous alignment. Thankfully my Monolith Flat is on the way, and I'm struggling to resist (but not too hard!) buying a DE1+ when they roll out again.

baldheadracing wrote:Perhaps follow what the manufacturer does? (You have to click on the graphic below to get to Reiss' Vimeo channel)
<video>

Thanks for posting this video. My first attempts with the Londinium tool were tragic. Watched the video and started to incorporate tighter circles around the perimeter of the basket while minimizing the more sweeping circles. That made all the difference.