Flow too low, as bad as too high?

Need help with equipment usage or want to share your latest discovery?
User avatar
dsc
Posts: 1166
Joined: 17 years ago

#1: Post by dsc »

Hi guys,

a few days ago I fitted my Elektra T1 with a Synesso ruby jet which is 0.6mm and goes straight inside the solenoid valve on the group. I've removed the original jet installed in the group, opened the needle valve on the supply line fully and have been running the machine like that for two days now. The water debit is 80ml/10s which includes the hissing and spitting during the first few seconds of the flush. With the lower flow I had to change the grind setting a little bit, but it wasn't anything drastic.

I don't really know why, but with this slower flow it's damn hard to get a proper extraction. They all show signs of channeling (suddenly appearing bright 'spots'), I get sprites in almost all of them and the dreaded 'twister' cone. In the cup the body is very low, the shots taste watery and more bitter then before.

My thought was that with lower flow it would actually be easier to get a proper extraction, but it looks like I was mistaken. Anyone had a similar problem/s?

Regards,
dsc.

User avatar
HB
Admin
Posts: 21983
Joined: 19 years ago

#2: Post by HB »

Very interesting topic, I hope you continue to post your findings as you gain more experience with the slo-mo A3.

Just this past Friday, Bob Barazza, who bought the evaluation model, was considering adding a delay-on switch for a couple seconds of low-pressure preinfusion. My hunch was that it would improve the forgiveness factor slightly and degrade shot clarity ("muddy" shots). Since reviewing the Elektra A3, I've held the unsupported opinion that sometimes rapid pressurization can be a good thing (see Pressure profiles, preinfusion and the forgiveness factor for related discussion).

My guess is that ultra-even wetting and rapid pressurization are the keys to the A3's superior clarity (if I were to speculate why, I'll offer improved fines migration with subsequent enhanced front-loaded extraction as the causal factor). Your findings may help support or refute my hunch. :?
Dan Kehn

User avatar
Arpi
Posts: 1124
Joined: 15 years ago

#3: Post by Arpi »

Low flow may behave similar to low pressure. With very low pressure, water has more time to seek the weakest path. On the plus side, it puts your distribution skills to the test and eventually may help you improve your shot. I think that a very well prepare basket tastes better but takes more time and carefulness.

Cheers

milledge
Posts: 5
Joined: 15 years ago

#4: Post by milledge »

This doesn't make sense to me. I don't see how low pressure would lead to increased channeling. Under low pressure situations the puck will have more opportunity to absorb water and expand. This would reduce channeling - not increase it. My understanding is that one of the advantages of the E61 machines is the progressive pressure ramp up/preinfusion. This preinfusion is reported to correct (minor) deficiencies in distribution. Why would low flow behave differently?

User avatar
dsc (original poster)
Posts: 1166
Joined: 17 years ago

#5: Post by dsc (original poster) »

Hi guys,

I got some fresh beans today and did some standard extractions from a double basket. I'm not sure why but I have to grind finer to get a proper extraction. This comes as a bit of a surprise as I thought that I would have to grind a bit coarser due to lower flow, but it turns out to be the other way around. I'm guessing that the low flow indeed acts a bit like preinfusion wetting the puck before full pressure is applied. On the Elektras you can actually hear the PF filling with water and the moment it reaches full pressure. There's a noticeable 'sssssssssssssssssssssssss-p' sound (yes that it proper engineering talk:)) where the '-p' means it reached full pressure. Comparing to the original jet it's much slower, heck it's even much slower then the smaller Kehin jet I got from Nicholas.

Taste wise I can't really say as I'm working with a blend that I haven't had for a while. The first shot from it was a gusher, but still tasted quite well, the others looked better but lacked in the taste department. I still get a bit of 'twisters' which I think is due to uneven distribution of fines during the initial low pressure wetting phase. Can it be that with full 1s ramp up the fines are simply pressed straight down, where as with low pressure ramp up the fines migrate creating 'patches' where most of the low-pressure water went?

Anyways I will keep using this jet for a while and then try to switch back to the small Kehin jet and compare the results. I still prefer the faster rampup and I'm not sure the pre-wetting phase is so cool. Well not on my Elektra anyway.

Regards,
dsc.

User avatar
dsc (original poster)
Posts: 1166
Joined: 17 years ago

#6: Post by dsc (original poster) »

Hi again,

thought I would post a quick update.

I've managed to nail the dose (14.5g) and grinder setting and I can now pull a pretty decent double shot. The extraction starts a bit funny as it's a tad too fast during the beading phase and slows down afterwards with the fines probably being pushed down under full pressure. The first few seconds is mostly drips and cone forming, but around the 13-15s I get signs of channeling in the form of a 'disappearing blond spot', the one that shows for like a second or so and then heals itself. The 'spots' appear in various places around the basket even when using WDT.

Now this might be all my fault, but I believe the Kehin jet was showing less signs of channeling.

Regards,
dsc.

User avatar
uscfroadie
Supporter ♡
Posts: 1156
Joined: 16 years ago

#7: Post by uscfroadie »

Arpi wrote:Low flow may behave similar to low pressure. With very low pressure, water has more time to seek the weakest path.
:?: :?: low pressure = channeling? I don't think so.
Merle

User avatar
michaelbenis
Posts: 1517
Joined: 15 years ago

#8: Post by michaelbenis »

This mirrors some of my findings with the open boiler lever machines vs. the closed boiler designs, where the water during preinfusion enters under gravity only in the former, but under pressure in the latter. I wonder whether the infusion is slower with lower pressure and therefore the grinds expand slower, which is why you are finding more channeling?

Cheers

Mike
LMWDP No. 237

User avatar
Arpi
Posts: 1124
Joined: 15 years ago

#9: Post by Arpi »

uscfroadie wrote::?: :?: low pressure = channeling? I don't think so.
Hi

At lower pressure, the distribution is more critical than at higher pressures. At very low pressure, the puck soaks at the rings before everything else. My findings correlate the opinions of this thread How to Preinfuse; Extraction Pressure Redux in terms of extraction unevenness at the ring. You can 'fix' that by paying good attention to basket preparation or by changing the technique. A good technique for high pressure may give poor results at low pressure because distribution defects are amplified.

Cheers

User avatar
dsc (original poster)
Posts: 1166
Joined: 17 years ago

#10: Post by dsc (original poster) »

Hi guys,

because the gicleur restricts flow the ramp up is slower and it takes around 4-5s before the puck actually gets hit by 9bar (well 8.7bar to be precise). It might be true that the puck simply absorbs some of the water and the situation is a bit like turning on the pump for a few seconds, turning it off for a few seconds and turning it back on. You won't get a good result from something like this.

It seems to me that restrictor is not a good way to limit water flow as it also affects pressure ramp up. The best thing would be low flow but at higher pressure, so you can get around 80-90ml per 10s, but get a 2-3s ramp-up to 9bar. This of course involves the pump and switching to a lower-rated model or using a TMFR pump at a lower speed (but also a low rated model). Is it only me or are rotary pumps slightly shite in espresso machines?

Regards,
dsc.

Post Reply