Great interesting work here. Nicely done.

Mark

I second mhoy's comment, well done.
I also run PI in my Silvia only I use a second OPV running off a bypass solenoid, it works a treat.

Just found this conversation when googling to see whether anyone else was doing the same as what I'm playing with.

I have a Gaggia Baby Class on which I'm controlling power to the pump by energising the coil for only part of each mains half-cycle. So a slightly different approach. I suspect your response curve is much more linear than mine, but mine has a greater resolution and, I imagine, a smoother flow.

Anyway, I thought I'd report that combined with a flow meter I'm now experimenting with PID controlling the flow of the entire extraction. Which means I get the same rate of flow regardless of grind and tamp, and also means I don't need an adjustable OPV. (The Baby Class's OPV is hard-wired to 11Bar as far as I can tell).

Results are variable so far - it's beginning to look like I need a different brew temperature now that I've reduced the flow rate. But I did get a very fine shot this morning, so I'm hopeful.

People have experimented with dimmers and SCrs. They found you can throttle the pump down to about 4 bar. Below that, there's no flow. I haven't heard of anyone doing it long term; so I'm not sure there's any info on how it affects pump life.

An alternative is to upgrading the valve on the steamer to a needle valve, and then to bleed water through it while making espesso. This works like a fully and rapidly adjustable OPV. Not as neat or easily automated as motor speed controls, but perhaps having a wider adjustment range.

Both approaches require an inline pressure meter, since the responses are quite non-linear.