another_jim wrote:There seem to be two effects on both machines. The first is to reduce or eliminate "outlier" shots. Presumably, this is because lower water mass reduces overshoot, or gets the boiler back to its base state more quickly. The second effect is the head scratcher: your lower boiler fill curves have a flatter temperature profile.

If I hadn't wasted too much of my life watching the displays on dataloggers while pulling sham Scace shots, I'd agree with you, but it is no head scratcher for me.

I've noticed this behavior before but not all the time and I think I understand it. It (presumedly) occurs when the boiler fill is fairly low, although in an autofill system it can appear to be random since you don't know what the boiler fill level is in an autofill system until you empty it.

What is going on is that at some point during the shot, generally around mid shot in my experience, the heating element has kicked in to the extent that it can heat the water and steam surrounding the heat exchanger, to the point where there is no further heat loss for the rest of the shot.

It is hugely more likely that this will happen in a HX system with low boiler fill, as the element doesn't have to squander its resources trying to heat a whole lot of water, rather it only has to heat a little bit of water and a lot of steam, which surround the HX tube and result in this observed shot temperature dynamic.

ken

It's a couple of weeks since I've updated this thread, and I just wanted to add a couple of additional observations that may be of interest.

My initial attempts to replicate the findings I'd made on my pourover vibe machine, in my plumbed in rotary machine, were stymied by two factors. One was that the lowest regular boiler fill level, with the stock autofill probe, was not a whole lot lower than the midlevel probe setting I was using before. At this setting the boiler had between 21 and 26 oz of residual water that I could drain out the water wand on shutting down the machine. This contrasted with about 8oz with the low fill setting on the sightglass of the vibratory machine. In addition, the boiler fill level in the autofilling rotary machine varied considerably, apparently by the length of time since it had last had its boiler drained.

I took the latter observation, of increasing boiler fill levels, as indirect proof of a leaking input solenoid, a problem I had once before, about 3 years ago, with this machine.

So, I replaced the input solenoid a week and a half ago, and also bent the autofill probe as far as I could to limit the resulting boiler volume. Now, the boiler volume that can be drained out the water wand is a consistent 14 oz. Although this is more than on the vibe machine, the boilers in the two machines are different as is the point where the water wands insert in their respective boilers, so I can't say that there is a difference in fill level based solely on knowing that 14oz can be drained in one machine and 8 oz in the other.

What I can say is that the rotary machine now has a consistent fill level that is much lower than it used to be. I have recorded many shot series with my Scace Thermofilter and datalogger, both "walk ups" and consective shots pulled every 2 minutes. The shot curves are much more consistent than before, and resemble the results I've previously posted in this thread from the pourover vibe machine, whose manually filled boiler is much easier to work with.

Here are a couple of examples:



and:



In summary, if you have put electronic temperature control in the boiler of your espresso machine, the boiler fill level is a very important variable that at the least must remain constant, if repeatable results are desired. Further, given a choice, you will probably find that your best and most consistent results (with regard to repeatable shot temperatures) will come from having your boiler fill level set at the lowest reasonable point consistent with safe operation of your machine. If you have autofill, you should make sure that your autofill probe is set to accomplish this goal of low boiler fill, and the probe needs to be clean without mineral deposits. If you notice that in spite of your efforts that your results are not good and that your boiler appears to be filling up with water, check the input solenoid if your machine is plumbed in, and replace the solenoid if necessary.

ken