I began with the understanding to source an old-school 600W dimmer - which I did - but through all fault of my own I blew it up. Oops. With my Gaggia's guts now hanging out and the hardware stores closed, I spotted a tabletop dimmer switch and decided it had a higher purpose than fine-tuning my lava lamp. It worked!

It's a Leviton TBL03-10W, costs $12:



It's only rated to 300 watts, but no worries, the pump in the Gaggia (and I suspect many other espresso machines) draws just 50 watts - and it only runs for 60 seconds every 24 hours - not a hard life for a dimmer.

It's dimmer-mod-hacker-friendly thanks to the included power cord. Just clip off the power plug, route the cord through the Gaggia's back vent slots, and splice into one of the power wires of the pump. Done!

The useful range of the slider seems pretty workable to me: about 1/3 of the way is about right for preinfusion, and the top third of the travel range covers the range I need for pressure/flow profiling. That may not seem like very much actual movement in the switch, but it's smooth (grease damped) and not twitchy. Plus, sliding a switch back and forth feels more like pulling a shot than turning a rotary dial.

The small form-factor is ideal for building into the Gaggia itself, which I combined with the PID controller:



The dimmer housing bolts right onto (with a little hacking) a 3D printed replacement for the back section of the top cover:



So the temperature controller and pressure profiling mods can be built back into the Gaggia - and still retain the upper water inlet:



CAD files are here: https://grabcad.com/library/gaggia-pres ... rol-deck-1

Looks good! Be aware that if it is a phase control (TRIAC) dimmer the pump (which is an inductive load) causes back EMF that could cause the dimmer to fail prematurely. It could fail open, causing the pump to stay on and eventually fail by exceeding its duty cycle - but most likely that this would happen as you are pulling a shot and you could simply unplug the machine.

nisb wrote: It could fail open, causing the pump to stay on

The pump and dimmer are still powered by the pump's rocker switch... The dimmer failing won't somehow draw power through an open rocker switch.
And if the Triac failed open, it wouldn't pass current, either, but now I'm being pedantic, we knew what you meant.

I concur about the switches still having authority over power delivery. Ultimately the potential for disaster will be contained by the local fire department when they hose down the charred remains of my home. And when my neighbors ask me why, I can smile and tell them "I was making espresso."

In the photo you can see I rewired the pump and 3-way valve to have their own red-glowy switches, so I can turn the pump off and on without releasing the pressure to the group head - works great for preinfusion.

As for the back EMF, I hooked up a (battery powered) oscilloscope across the dimmer circuit to see just how bad it gets:



Hits about 190 volts - about 60% over line voltage. I can't read the markings to confirm, but I bet most Triacs are manufactured to play in worldwide markets and as such have breakdown voltages above 250 volts.

Update: I hacked my hack.

Bridged the slide potentiometer in the dimmer with a 270K resistor:



This lops off the bottom third or so of the dimmer's adjustment range, which wasn't really doing much. Now the entire range of adjustment is useful. When turned all the way down I get a low-pressure pre-infusion that hits around 4 bars (measured static) and saturates the puck in around 5 seconds. At that point I can slide to full pressure, or if I'm in the mood for a little more hang-time before starting the 'pull' I can hit the pump switch (off) while leaving the 3-way switch closed.

So, my question: how important is the gauge - has there been a magic moment in pressure profiling that couldn't have happened without the gauge?

Replying to a question in another thread.
I swapped the temperature gauge in my Sirena for a pressure gauge, knowing that I would install a PID for temp monitoring. I find the pressure gauge helpful and entertaining as I enjoy seeing how everything is working in real time. The analog gauge can help you do some manual profiling by providing live feedback while you are manipulating the pump via the dimmer. It's fun to play with, but in the end, manual flow manipulation may not be a big part of your workflow.

As for placement, it looks like you can easily reprint your mounting bracket there and put the gauge where that vent is. This is pretty common with those GCP Modbox setups. If your plan is to incorporate flow via bluetooth scale, it might make more sense to use a transducer instead and not worry about mounting a gauge.

bgnome wrote:... manual flow manipulation may not be a big part of your workflow.

As you aptly spotted, the vents in my cover plate are a placeholder for a pressure gauge. Is 'Modbox' a thing, or a blanket term for modded GCPs?

Adding the dimmer has rescued what otherwise would have been a sink-shot when I get my grind setting wrong, so I'm happy to have added it. Espresso has so many ways it can go wrong, *sigh*. The dimmer is kinda the last chance for recovery before having to reach for extra sweeteners.

I've been adjusting the dimmer to maintain a nice dark color over the span of about 30 seconds. If I had a pressure gauge I don't know what circumstance would lead me to let the shot blonde or mess with the extraction time just for the sake of pressure - hence my question if I'm missing something. That said, I read a post by Decent Espresso that their preferred profiles are pressure-based...

I think it's more about experimentation with different profiles. To paraphrase James Hoffman, it won't necessarily lead to better espresso, just different espresso.

For me, I'm not really interested in profiling. But as an engineer, I love to be able to quantify and measure things. If you are anything like that, FOMO will likely lead you in a certain direction anyway. If you are satisfied with your method, I don't think you really have anything to gain from being able to visualize pressure.

bgnome wrote:For me, I'm not really interested in profiling.

You made me think about what I've been doing - thank you!

Yes, I'm using the dimmer mod - but not to profile. I'm using it more like a faucet, and I adjust it a little differently each time. Variations in the beans, the grinder, how I stir, level, and tamp all seem to be key drivers in the shot outcome that will smash my attempts to follow a predetermined flow/pressure profile.

Based on my history in semiconductor manufacturing, I consider making espresso to be an "out of control process". There are at least three reasons if a shot runs too fast - a nightmare scenario for a process engineer This has broken my will as an engineer - which may be why it's so fun!

Looks like this is a forum chock-full of engineers. MSEE here, RF/microwave design. Used to work in RF semiconductor like GDM, back when I was pursuing a Ph.D., but got fed up with it and left.

I love the mod, especially the packaging of it. In fact, it'd be interesting to hang the scope on that pump again, now with the 270k bridging what I assume is the output of the triac. I suspect it may lower the peak voltage for the same pump speed, but that's based more on just reasoning the fact that you're now applying voltage during the off cycle of the triac, I have no actual experience with AC dimmer circuits.