Hey all — I've been sitting on this experiment for a couple weeks and finally compiled everything into something readable. I know pressure profiling gets talked about a lot here but I wanted actual data rather than vibes, so I ran a fairly controlled test on my Gaggia Classic Pro with an aftermarket pressure gauge kit and an OPV adjusted to both 9 bar and 6 bar.

Setup: 40 total shots — 20 at 9 bar, 20 at 6 bar. Same bag of beans (medium roast Ethiopian Yirgacheffe, washed process, about 2 weeks off roast). Same 18g dose, same 36g yield target, same grind setting throughout (I'll address this in a moment). I used my Acaia Lunar for weight and a VS Poco refractometer for TDS. Shots were distributed using WDT and tamped at roughly 15 kg using a calibrated tamper.

Here's the summary data:

Pressure	Avg Extraction Time	Avg TDS	Crema Quality	Flavour Notes
9 bar	27.4 sec	9.2%	Thick, dark, persistent	Bittersweet, stone fruit, slight astringency on finish
6 bar	31.8 sec	8.8%	Lighter, fades faster	Cleaner cup, brighter fruit, mild acidity, less crema

The 6-bar shots were noticeably slower, which makes sense — less pressure means the water takes longer to push through the puck. What surprised me was that the taste difference wasn't as dramatic as I expected from reading threads here. The 9-bar shots had more crema and a slightly fuller body, but also a mild bitterness on the back palate I didn't notice as much at 6 bar. The Ethiopian fruit notes were more distinct at 6 bar — cleaner, more clarity.

The TDS difference is small (9.2% vs 8.8%) but consistent. Worth noting I kept grind constant intentionally to isolate the pressure variable — I know that's not ideal for real-world comparison but it removes a confounding variable.

Anyone else run something similar? Particularly curious whether others noticed the clarity difference on lighter roasts. Would love to try this same test with a dark Italian blend next.

This is excellent methodology — I really appreciate that you actually used a refractometer instead of just saying "tasted better." The discipline to keep grind constant is smart for isolation, even if it's not representative of optimal espresso at each pressure point.

Your results match my experience exactly. On my Lelit Bianca I have the flow paddle, which lets me do essentially anything between 0 and 9 bar. I've settled on a 4-bar pre-infusion phase for 8 seconds followed by a ramp to 7.5 bar, and the Ethiopian-style light roasts I favour genuinely sing at lower pressures. The fruit clarity you're describing is real — it's not placebo.

One question though: did you make any grind adjustments between the 9-bar and 6-bar shots? Even if you kept it constant for experimental consistency, I'm curious whether you noticed the shots running looser at 6 bar (channeling more?) given the same grind. At lower pressure the water resistance dynamics change meaningfully.

Also — what grinder are you running? That might matter for how cleanly the lower pressure shots were expressing.

Solid data. Going to push on one methodological point though because it's important for anyone interpreting these results.

When you change extraction pressure, you should compensate with grind size to maintain the same target flow rate. The reason is that espresso extraction quality correlates strongly with flow rate — approximately 1–2 ml/sec is considered ideal by most practitioners. At 9 bar with your grind, you'll be achieving a certain flow rate through the puck. At 6 bar with the same grind, the flow rate drops because there's less driving pressure. Your longer extraction time at 6 bar (31.8 vs 27.4 sec) confirms this — you're extracting slower, which confounds the pressure variable with a time/rate variable.

In a "real world" comparison where a barista has dialled in each pressure separately, you'd grind finer at 6 bar to compensate — which would bring the extraction time back toward ~26-28 seconds. This might actually produce a higher TDS at 6 bar, not lower, and the flavour clarity could be even more pronounced.

I'm not criticising the test — for isolating pressure as a single variable you did the right thing. Just important context for anyone drawing "9 bar vs 6 bar" conclusions from this without understanding the grind relationship.

What grinder are you using, out of curiosity?

Good catch from both Elena and GrinderGuru — yes, I deliberately held grind constant to isolate pressure as the single variable. It's a controlled experiment, not an optimised comparison. You're absolutely right that in real practice you'd dial the grind to compensate for the pressure change and bring flow rate back into target range.

Grinder is a Niche Zero, for the record. I'm around setting 17 for my usual 9-bar pull. Didn't change it for the 6-bar series.

The follow-up I want to do is: start from scratch at 6 bar and properly dial in grind to match a ~28-second extraction time, then compare TDS and flavour to a properly dialled 9-bar baseline. That would be a more "apples to apples" flavour comparison even if the grind settings differ.

EspressoElena — I didn't notice significant channeling at either pressure in this batch. The puck was intact on inspection. I think the Niche's consistency helped. But you raise a valid point about lower pressure and channeling dynamics — I'll keep an eye on that in the follow-up series.

This mirrors my exact journey. I modded my Rocket Appartamento last year — adjusted the OPV from 9 bar down to 7.5 bar and the improvement in clarity was immediately obvious. Didn't go all the way to 6 because I was nervous about losing too much crema, but even 7.5 was enough to notice a real difference on washed Ethiopians and Kenyas.

I never went back. Genuinely one of the easiest upgrades I made and it cost me about £15 in parts. If you want a visual walkthrough, this YouTube video covers the Appartamento and Classic Pro OPV mod really clearly: youtu.be/espresso-opv-mod-guide (fair warning it's 22 minutes but worth it).

Your data showing lower TDS at 6 bar tracks — but I wonder how much of that is the slower flow rate vs. the pressure itself. GrinderGuru's point about compensating grind is exactly right. Once dialled in properly I'd bet you'd see TDS closer to parity, just with different flavour character.

Decent data, genuinely. But I'll be the voice of dissent here: 9 bar is the industry standard for a reason. The majority of commercial espresso blends — the ones used in 95% of cafés worldwide — are roasted and developed specifically for 9-bar extraction. The roast chemistry, the expected solubility curves, the oil migration that creates crema — all calibrated for 9 bar.

The "lower is better" crowd is almost exclusively pulling single origin light roasts. Absolutely nothing wrong with that, but it's not the same as espresso in the traditional Italian sense. Try pulling a Lavazza, an Illy, a Segafredo blend at 6 bar and see how hollow and under-extracted it tastes. The heavier roast, the more robust blend — 9 bar is exactly what it needs.

I'd genuinely love to see this same experiment run with a traditional Italian espresso blend at both pressures. I suspect the 9-bar results would look better in that comparison. The Ethiopian washed result you got is interesting but niche (pun intended, sorry Benny).

Not dismissing the test — more data is always good. Just pushing back on the implicit conclusion that lower pressure is generically better.

TamperKing makes a fair point and I mostly agree — but I'd frame it slightly differently. It's not that 9 bar is "better" or lower is "better" — it's that optimal pressure is roast-dependent.

I've done extensive tasting across different roast levels and the pattern I see is consistent:

• Light roasts (washed, natural): lower pressure (6–7.5 bar) tends to give better clarity, less bitterness, more terroir expression
• Medium roasts: anywhere from 7–8.5 bar, very bean-dependent
• Dark roasts / blends: 9 bar (or close to it) — more body, better crema, the roast flavour is what you want and it needs that pressure

So yes — I'd love to see this same test with a dark Italian blend, as TamperKing suggests. My prediction: the 9-bar result wins clearly on a dark roast, the difference inverts vs. what PressureProfiler found here.

PressureProfiler, have you tried any medium roasts at the different pressures? Would be interesting mid-point data.

Let me geek out on the physics here because I think it explains everything you're both seeing and arguing about.

Pressure, flow rate, and channeling — the fundamental relationship:

Espresso extraction depends on water moving through a bed of coffee particles. The flow rate is governed by Darcy's Law: flow rate is proportional to pressure and inversely proportional to the resistance of the puck. When you raise pressure, you increase the driving force — but you also increase the shear stress on the puck, which can cause channeling (localised paths of least resistance where water flows through without extracting properly).

At 9 bar, channeling risk is higher — particularly with inconsistent puck preparation, lighter tamping, or certain grind distributions. A channeled shot will have higher TDS from the channeled zones but under-extraction overall — which is why 9-bar shots can simultaneously taste stronger and more bitter (over-extracted channels) while being less nuanced (under-extracted majority of the puck).

At 6 bar, the driving pressure is lower, so the water moves more slowly and more evenly through the puck. This reduces channeling risk, which means more even extraction. That's likely accounting for a significant portion of the "clarity" effect PressureProfiler noticed — not the pressure per se, but the more even extraction that lower pressure enables.

This is also why pre-infusion at low pressure followed by higher peak pressure is potentially the best of both worlds: you saturate the puck evenly first (eliminating dry spots and reducing channeling risk), then ramp up pressure to drive extraction. It's what flow control machines like the Decent DE1 and Lelit Bianca allow you to do manually, and it's the core value proposition of pressure profiling.

The conclusion isn't "lower pressure is better" — it's "more even extraction is better, and lower pressure is one way to achieve it."

Really enjoying this thread as a newcomer — this is exactly the kind of discussion I joined BaristaChat for.

Quick question though: does any of this apply to home machines in the entry-level range, or is it mainly relevant for prosumer setups? I have a Breville Barista Express and it runs at 9 bar out of the box. I've seen people mention an OPV adjustment mod for it — is that something worth doing? Or is it only meaningful if you're running a higher-end machine where you can actually taste the difference more clearly?

Apologise if this is a dumb question — happy to be redirected to another thread if there's already a dedicated guide for this.

This thread has gone better than expected — genuinely appreciate the quality of discussion here. Let me do a quick round of responses:

GrinderGuru & Elena: Yes, follow-up test will include grind compensation to match flow rate at both pressures. That's the more honest flavour comparison and your methodological critique is 100% valid.

TamperKing & CuppingQueen: Totally on board with the roast-level framing. The dark Italian roast test is on the list. My intuition agrees that 9 bar will perform better there — I'm genuinely curious whether the inversion is as clean as CuppingQueen predicts.

MicrofoamMike: That's one of the clearest explanations of channeling physics I've read. The Darcy's Law framing is useful — bookmarking that post for future reference. The pre-infusion point lands perfectly, and I'm planning to experiment with a proper profiled pull (low pressure saturation → ramp) once I have a better gauge setup.

ShotClockSam: Not a dumb question at all, glad you asked. Yes, the OPV mod is absolutely worth doing on the Barista Express — it's one of the most popular mods in the community for exactly that machine. The stock 9 bar is noticeably high for light-to-medium roasts. There's a detailed guide with photos in the Equipment section: Breville Barista Express OPV Mod — Complete Guide (with photos). Takes about 20 minutes, minimal tools.

I'll post the follow-up test (with grind compensation, dark roast, and a profiled comparison) in the next few weeks. Will link back to this thread. Thanks everyone.