I've got a Quick Mill Alexia Evo which overall I love very much. But the heater seems to "blip" on and off a little quickly for my tastes once the machine is up to temp. I've got no issues at all that I know of with it getting to temp, overshooting, or anything like that. I pull a shot a day, each morning, after about a half hour warm-up. Second shots and other recovery needs don't really exist.

I know that, since it works, I should leave it alone. And I have, for a few years now other than cleaning and such. But it turns on the heater for about a half second, off for about a full second (these are guesses), and repeats until I make my shot, flush it a bit, and turn it back off until the next morning. This constant blipping has the effect of a light on the same circuit dimming on and off, and overall makes me wonder if I can "wear out" the PID this way?

I'm set to Fahrenheit
P = 1.0
I = 0.01
D = 1.5

I've read and read and searched but am just lost on what to do. I don't need an overhaul of settings, I don't think. But I'd like to slow this guy down a bit. Seems that once it's up to temp, it doesn't need to blip every second or so to maintain it. I don't have a time frame in mind, but I'd think... 5 seconds, 10 seconds, maybe 30 or even 60 would be more appropriate?

Is there a particular setting change that anyone recommends that seems like it should calm things down a little bit?

You could get a machine with a traditional "Klixon" thermostat or a pressostat and a wide deadband, but the whole purpose of the PID is to do exactly what it's doing. Yes, anything can and will eventually wear out, but assuming the SSR is reasonable for the machine, it will be after several hundreds of thousands of cycles.
As for the light, you might want to give the machine its own dedicated 15 amp outlet. That circuit is stressed.

Worth unpacking what a PID is and what is going on.

But first the SSR. Solid State Relay. These are usually a Triac, which is an AC device (it is in the name) which switches in an interesting fashion. Trigger it on, and it conducts current the the direction that the potential (aka voltage) across it is facing. When the voltage drops back to zero, or swaps direction, the device stops conducting. But is is built as a pair, and can be triggered to conduct the other way. The upshot is that the internals of a typical SSR are switching on and off with every half cycle of AC. So in the USA 120 times a second. (In the civilised world that would be 100 times.) The additional cycling of the entire device every few seconds is not a concern.

PID stands for Proportional Integrating Differentiating. Each of these terms has specific meaning.
The differentiating part allows the controller to take notice of how fast things are changing (ie the differential wrt time) and use that as an input into how it controls things. In a boiler it may see that it is closing fast on the set point, and reduce the amount of heat to avoid overshooting, or it may see a fast fall in temperature and use that to trigger full power reheating.
The integrating part allows it to take note of the sum of conditions over time (ie the integral wrt time). So maybe allow for stored heat in the heating element. The actual setting become a proxy for the dynamics of the system being controlled.

The proportional bit is also key. Unlike a simple thermostat, a PID wants to be able to regulate the amount of controlling input it has. So from very low heat input to full heat. This could be done in a few ways, and how partly depends on the system being controlled. A Triac can provide proportional control of AC - it is after all how light dimmers work. By chopping the sinusoidal waveform so that you get a variable amount of the cycle. If you were controlling say a light bulb you would need to do this. But a simple boiler is a much slower reacting device, so you can get away with multi second on/off cycles to achieve proportional control.

The nuance here is that the PID is not acting like a simple on/off thermostat, although from the outside it does look a lot like a tight deadband thermostat. A thermostat turns on and off, with some hysteresis, centred about the set point. A PID controller will start to throttle back and modulating the power before it has reached the set-point, sneaking up on the set point and avoiding overshoot. Something a on/off simple thermostat can't do.

I suspect the main reason for choosing to use a multi-second pulse width modulation versus a per AC cycle pulse width modulation - although they would appear to achieve the same result - is that AC cycle PWM causes RF interference, just like a light dimmer. So your machine is less likely to create annoying buzzes in the mains. A SSR will usually switch as the AC cycle crosses zero volts, which further minimises interference as well.

The PID is not the issue.
If a light on the circuit is dimming significantly, there is an issue with the electrical circuit wiring.

How old is the house?
Has the breaker panel been checked for corrosion, loose wires or tired breakers?

What else is on the circuit?

Voltage drop should be about 3vac over a 60 foot run and should not dim any lamp appreciably. You would need a photometer to detect it.

As it happens, I have the older PID'd model of Alexia. Great machine.
And yes the industrial PID is set to keep your boiler temperature very exact. Too exact for humble espresso makers like us.
Do you need that rapid flash? No but that's just how the programming was set up.
My Watlow PID has a parameter called Time Base. It can be set to very rapid to any number of seconds.
I suggest you look at your PID manual to see how to set up the time base to a number you can live with. I think 10 seconds will work. The thermal inertia of the machine will even out any on/off variations. Failing that parameter you could play with a wider P value. But tuning PIDs is a rabbit hole to itself.
As far as dimming lights go, I tried my Alexia in every house outlet and the 12" utility outlet at the breaker panel allows normal machine and grinder operation, with no lights dimming anywhere. Nice to have after 10 years of dimming lighting. So I now have a workshop espresso machine, and a rather dumb new machine in the espresso corner.
Dave

This certainly took an interesting turn but you guys may be right here.

House was made in the 90's. It passed inspection about 8 years ago when we bought it and the inspection felt thorough but I am sure the breakers weren't scrutinized nor the outlets checked for more than GFIC if near a sink. That particular circuit does sort of traipse around the basement but little else ever runs on it, especially at the same time.

I've been in the panel a few times for other work and things appear good, but it's entirely possible the breaker itself could need replacing. Hmm.

Davi-L wrote:I suggest you look at your PID manual to see how to set up the time base to a number you can live with. I think 10 seconds will work. The thermal inertia of the machine will even out any on/off variations. Failing that parameter you could play with a wider P value. But tuning PIDs is a rabbit hole to itself.
Dave

I see little in the manual regarding the PID and no information on the PID itself. Any way to tell without opening the machine and hoping I can find a brand and part number?

It's about 3 years old, and visually looks identical to the product on this page (in fact - I recall this is where I bought it): https://www.totalespresso.com/products/ ... alexia-evo

I do NOT plan to ignore the wiring question, I'm a fan of safety. I would also like to reduce this cycling time a bit if easy enough to do so.

Ooh, you have the QuickMill factory PID. The user manual would not cover all the parameters. Yes, you need to get the exact model and then try to download the hard core PID manual.
My PID is a generic USA model.
Regarding house wiring. My newly wired house met all the electrical codes and was done professionally. I watched. However the Electrical Codes do not take into consideration the use of industrial PIDs on espresso machines.
It comes down to gauge of wire and distance. 18" of 14 gauge at my breaker panel powers the Alexia quite well. Any further and the PID makes the lights dim annoyingly. All 12 electricians I spoke to about this matter had no idea what I was talking about. They don't PID their drip coffee machines.
D.

I'll check out my breakers and wire gauges, and connections at each end. I may pull off the cover and find more details on the PID.

I admit I was also sort of hoping to change the PID settings with something like - "change P to 70% of what it is set to now and that should slow it down". Maybe not that, but something along those lines. Seems it's not that simple.

The PID doesn't care how fast it switches. There is nothing mechanical in a PID to wear out. The only issue I can think of with a PID that switches fast is if you use the wrong kind of SSR, one that only turns off at zero crossing, it will effectively turn on the power for longer than it expects and likely cause it to become very confused.