(This is Crossposted on home-barista.com; if you choose to comment please comment on one but not both of these crossposted threads)
Due to questions raised on a thread on home-barista.com, the issue of the impact of autofill on shot temperatures and pressure was discussed. I have started this thread as a repository for any actual observations that might be made on this topic.
The original issue raised was what would be the impact on a shot if the autofill was actuated during the shot itself, with particular reference to vibratory pump machines with autofill. There are two issues here, the first being the effect on the shot pressure, and the second being the impact on shot temperatures for both the shot in question and for subsequent shots in a consecutive shot series.
I do not have a vibe pumb machine with autofill, although I do have one without it. Testing that machine would probably not produce useful information as the boiler is filled manually. I do have a rotary pump driven plumbed in machine, a Cimbali Junior D model which has been modified with both a PID for more precise boiler temperature control, and a (defeatable) delay timer for inducing preinfusion using regulated water mains pressure of about 3 bar for the initial 6 seconds of each shot. Because of concerns about the delay timer confusing these issues, I elected to bypass the delay timer (it is switched in my installation) leaving the machine in its native configuration other than for the PID.
As the first part of the test I installed the portafilter (PF) manometer and tested shot pressure, which remains as previously set at 9 bar. Then, as a second test I again tested the shot pressure with the PF manometer but after 10 seconds opened the water wand and dumped out 200ml of water which actuated the autofill almost immediately. The shot pressure did not change and remained at around 9 bar. Therefore, with at least this rotary pumped machine, shot pressure was not effected by the autofill coming on during the shot.
The next thing to test was the impact of autofill actuation on measured shot temperatures, both during the shot and in consecutive shots afterwards. I started with a PID boiler temperature setting that I normally use when I'm making milk drinks with the Harrar Horse SO beans I have recently been using for espresso. The Scace Device and an Omega Datalogger were used to record shot temperatures.
As a baseline I performed a 4-shot series without actuating autofill to show the consistency of shot temperatures observed:
Having obtained a baseline, and waiting 45 minutes to allow the machine to return to baseline condition, I performed a six shot series with autofill activating during the 2nd shot. The first shot was pulled, as before, after an initial 50ml cooling flush. Ten seconds into the 2nd shot, I drained 100ml of water out of the water wand while continuing to record the shot. The autofill ran audibly during the 2nd half of this shot and continued into the idle period in between the end of the 2nd shot and the beginning of the 3rd. The shot series continued as before in the baseline series, with 1 shot pulled per 2 minutes, equating to a pause of 90 seconds between the end of one shot and the beginning of the next one:
As you can see, the average shot temperature in the baseline series is a bit higher than that observed in the series where the autofill was actuated. The 2nd shot, during which the water was drained from the boiler and during which autofill was running the last half of the shot, did not appear to have its shot temperature effected by this maneuver. There was a trend towards declining shot temperatures with each subsequent shot, however within the range I have previously observed in this machine as set with this boiler temperature and without the autofill being actuated.
In summary, in my rotary pump driven Cimbali Junior D HX machine, I have observed the following with boiler autofill actuation during a shot series: (1) no effect on measured shot pressure during the shot when autofill was operating, and (2) no effect on shot temperature during a shot in which autofill actuates. Effects of autofill actuation that I DID observe were a trend towards overall lower shot temperatures in the shot series, especially in shots FOLLOWING the shot pulled simultaneous with autofill, and a fairly limited effect overall, all things considered, given the predicted pertubation of the system that I anticipated.
Finally, this is not a test of a vibe pump machine with autofill, and individuals owning such machines will need to test them in order to see how generalizable these observations might be.
> Due to questions raised on a thread on home-barista.com, the issue of the > impact of autofill on shot temperatures and pressure was discussed. I have > started this thread as a repository for any actual observations that might be > made on this topic.
Ken, Nice piece of work! I've often wondered why mfgrs don't blockout autofill when a shot is being pulled? It happens so infrequently, that I suppose it is no big deal. What I do know is that when it happens to me, I toss that shot.
My machine has a rotary, triple diaphram pump. It is somewhere inbetween a vibe pump's and rotary vane pump's performance. It has no problem keeping up doing both chores.
However, autofill on my machine means that the boiler pressure drops to near zero, less than 0.1 bar. If autofill occurs during a shot, I wait 5 minutes before trying again, otherwise everything is too cool. Dan
>> Due to questions raised on a thread on home-barista.com, the issue of the >> impact of autofill on shot temperatures and pressure was discussed. I >> have started this thread as a repository for any actual observations that >> might be made on this topic.
> Ken, Nice piece of work! I've often wondered why mfgrs don't blockout > autofill when a shot is being pulled? It happens so infrequently, that I > suppose it is no big deal. What I do know is that when it happens to me, > I toss that shot.
> My machine has a rotary, triple diaphram pump. It is somewhere inbetween > a vibe pump's and rotary vane pump's performance. It has no problem > keeping up doing both chores.
> However, autofill on my machine means that the boiler pressure drops to > near zero, less than 0.1 bar. If autofill occurs during a shot, I wait 5 > minutes before trying again, otherwise everything is too cool. Dan
Hi Dan,
There are a couple of things you might try doing to reduce this annoyance. The first thing would be to carefully look at the placement of the autofill probe in your boiler. If the tip is not at the widest part of the boiler (e.g. in the middle from bottom to top) you could move it to that position. I have found in my machine that if the probe is a little bit above or below that point that it becomes more sensitive and actuates much more often, since the cross sectional diameter of the point at a higher or lower point is less and it will take small amounts of boiler volume reduction to trigger the autofill. Of course, not all autofill probes can be adjusted like that so if your's can't than this option does not apply to you although perhaps the probe has become bent and in that case it could be bent back to be positioned correctly.
The other possible thing to do is once a day or every other day to bleed off enough water through the water wand to actuate the autofill. That way the boiler will be filled as high as it is going to go on "the cycle" and small drops from frothing won't be enough to trigger the AF. In a sense you are beating the autofill to it, and this should prevent most possible AF actuations during a shot unless you are making Americanos with boiler water.
In article <G96dnQ3XXqoJWoLYnZ2dnUVZ_u6dn...@insightbb.com>, "Dan Bollinger" <danNObollin...@insightSPAMbb.com> wrote:
> > Due to questions raised on a thread on home-barista.com, the issue of the > > impact of autofill on shot temperatures and pressure was discussed. I have > > started this thread as a repository for any actual observations that might > > be > > made on this topic.
> Ken, Nice piece of work! I've often wondered why mfgrs don't blockout > autofill > when a shot is being pulled? It happens so infrequently, that I suppose it > is > no big deal. What I do know is that when it happens to me, I toss that shot.
I have a vibe pump, autofill machine. It being an HX (as Ken's) machine, for clarity, the above should only be an issue with single boiler (non-HX) machines, correct?
The autofill kicks in when the steam boiler level goes down, which understandably happens when I'm steaming. But when I'm pulling a shot, the pump is busy pushing water through the HX. There's no reason for the level to go down in the boiler... AND - there's only one pump. I'm not 100% on the water path - but how could one pump be pumping to 2 locations simultaniously? There MUST be a diverter valve that determines whether water goes through the HX or into the boiler.
Dan Bollinger wrote: >> Due to questions raised on a thread on home-barista.com, the issue of -snip- > However, autofill on my machine means that the boiler pressure drops to > near zero, less than 0.1 bar. If autofill occurs during a shot, I wait > 5 minutes before trying again, otherwise everything is too cool. Dan
As Ken said, I'd be concerned that your probe is misplaced, allowing a greater loss of water before the autofill kicks in, which then lowers the temp excessively. I'd also check that the probe is perfectly clean near the tip, so that it doesn't overfill slightly.
> I have a vibe pump, autofill machine. It being an HX (as Ken's) > machine, for clarity, the above should only be an issue with > single boiler (non-HX) machines, correct?
> The autofill kicks in when the steam boiler level goes down, which > understandably happens when I'm steaming. But when I'm pulling a > shot, the pump is busy pushing water through the HX. There's no > reason for the level to go down in the boiler...
Vibration. When the autofill is JUST ready to kick in, and you bump your machine inserting the PF, or the pump runs, this can cause ripples on the boiler water surface. The autofill can sense that it is no longer immersed and open the fill valve and turn on the pump. I've run my grinder and had it kick on my autofill.
> AND - there's > only one pump. I'm not 100% on the water path - but how could one > pump be pumping to 2 locations simultaniously? > There MUST be a > diverter valve that determines whether water goes through the HX > or into the boiler.
> What am I missing?
The solenoid valve for autofiling is off of a tee on the tubing on its way to the HX. At least in my machine it is.
> There are a couple of things you might try doing to reduce this annoyance. The > first thing would be to carefully look at the placement of the autofill probe > in your boiler.
Thanks, Mine is already set about center, otherwise my HX is not immersed and I'd have the same problem Ian had.
> The other possible thing to do is once a day or every other day to bleed off > enough water through the water wand to actuate the autofill.
Way too much maintenance investment for me. I'd be investing doing something daily in order to prevent something that happens twice a year. I could spend less time inserting a cut-off switch the next time I had the machine apart.
They say that 90% of life is maintenance of life. That leaves 10% for enjoyment. If I reduce my life's maintenance to 89%, that means I'd be increasing my time for enjoying life 10% !
>> However, autofill on my machine means that the boiler pressure drops to near >> zero, less than 0.1 bar. If autofill occurs during a shot, I wait 5 minutes >> before trying again, otherwise everything is too cool. Dan
> As Ken said, I'd be concerned that your probe is misplaced, allowing a greater > loss of water before the autofill kicks in, which then lowers the temp > excessively. I'd also check that the probe is perfectly clean near the tip, > so that it doesn't overfill slightly.
Thanks, and its depth is properly set. It probably needs to be cleaned, as you say, because I've noticed that it is running a little longer as time goes by. Cleaning it would increase its sensitivity and reduce deadband. Dan
In article <aK6dnWEv2I7Mc4LYnZ2dnUVZ_oudn...@insightbb.com>, danNObollin...@insightSPAMbb.com says... snip
> They say that 90% of life is maintenance of life. That leaves 10% for > enjoyment. If I reduce my life's maintenance to 89%, that means I'd be > increasing my time for enjoying life 10% !
Good point, but if you learn to REALLY enjoy bleeding hot water from your machine, you can have it all, maintenance AND enjoyment!
Ted
-- email me at: tee en jay ess eye em pee ess oh en one-the-number (at) cee oh em cee a ess tee (dot) en ee tee
ANY other email addie will probably mean I spam-killed your message unread, by accident, really.
