Strega-fying my Brugnetti Aurora, AKA a non-invasive way to add group heater to any lever machine

A haven dedicated to manual espresso machine aficionados.
Supporter ♡
Posts: 2483
Joined: 13 years ago

#1: Post by samuellaw178 »

Sorry for the cheeky title. :P Hopefully what I am going to share is worth your time (if you're a commercial lever user).


I've had my Brugnetti Aurora lever for almost 4 years now and is without a doubt the most gratifying machine I've ever used. But ever since I read about PID-ed cartridge heater on Strega, I am a convert of the concept and strongly believe that ANY modern lever should have that (for reasons expanded in this post).

As some of you may already know, on most lever machines, the actual brew temperature of a shot is largely determined by the heat status/temperature of the lever group head. The boiler temperature does have some impact on the brew temperature, but to a lesser degree (that's why PIDing a lever makes lesser sense and why the tiny fluctuation in pressurestat does not matter). You probably also know this from experience with the smaller levers (Pavoni etc with lesser thermal mass) - if your group head is overheated, the chance of getting the target brew temperature by adjusting boiler temperature is pretty much as good as a snowball's chance in hell. Vice versa if your group head is too cool.


There is nothing wrong with heating and driving the group head's temperature using readily available passive heat from the boiler. It is a simple tech that works. But the flip side is it also means you are stuck with a certain temperature profile. In doing that, you're essentially leaving the heat control entirely to the passive heat transfer(read - no control) from boiler (and possibly other external factors - air draft, the extra heat from brew water, ambient temp etc).

Allow myself to Indulge with my Aurora artwork for a second... :P This is my understanding of how the temperature of a lever group works:

I have been trying to incorporate the group heater feature onto my Brugnetti Aurora, but stumbled into two major roadblocks (i) I didn't want to make any permanent modification to this piece of history, if possible (ii) There is not enough metal/place on the group where I can insert a cartridge heater without interfering with the water path.

About two years ago, I have experimented with this silicone band heater (below). It worked to some extent, but not that great because the in-built thermostat is regulating the band heater's temperature, not the grouphead's. There is also a huge temperature band of +/- 10 deg C, and some other flaws. More importantly (deal breaker for me), it completely ruined the look of the lever (they are only available in the bright orange/red color). The band heater went into storage after the novelty wears off about two months, that is until recently...

Photo (credit from Ebay): You can find these on Ebay by searching for Silicone Band Heater. I am not advodating the brand but they are whom I used (hence the brand is censored). They can custom make the length/width to be pretty much any size you need.

One day, I was wondering why not PID the band heater and place a temperature sensor on the group head - the actual temperature that you want to control (instead of measuring the band heater temp by the stock probe). And so I undertook this small project to make it happen...

The Modification

Before I present the information....

Disclaimer & warning: I am not advocating anyone to do this. All information posted in this thread is only for entertainment purpose. If you do attempt to replicate anything from this post, it is entirely up to your discretion and at your own risk. If you are not an electrician or have sufficient electrical knowledge you can get yourself electrocuted or worst causing electrical fire/hazard.

With that disclaimer made clearly....What you will need are:
  • Silicone heater band custom made to the dimension of your group head (no need for fancy measuring tool, just a measuring tape will do)
  • PID, SSR, Bare K-type thermocouple, PID enclosure/aluminium project box
  • Stainless steel cable tie & crimper/plier (this is key as you will find out soon)

Here's the general concept/scheme -

Photo: You can either power the PID & heater directly from the machine, or by using a separate line/plug like I did so you can choose to turn the group heater on/off at will. Make sure the machine is properly grounded in both case.

This is before the modification:

This is after modification:

Photo: You probably couldn't even tell if I hadn't mentioned it! :p The temperature strip was added to monitor the group temp but at this point it was almost pointless since the temp is very stable and almost never show changes.

The silicone band has an adhesive back so it could definitely work as it came from the factory. However, the stainless steel cable tie solution was important for two reasons.

One, you definitely want a secured & full contact between the band heater and your group head for this to work. The cable tie is a more mechanical/permanent method to do that (instead of just relying on the adhesive which may come off with age/heat cycling). I've also managed to 'clamp'/sandwich the bare K-thermocouple in between the non-heating portion of the band and group surface. So there is no need to drill the group to mount the temp sensor (yeah I had considered that...).

Photo: A closer look

Second, it is for aesthetic reasons - I've tried a couple other methods to cover this band's orange/red ugliness, but this cable tie solution seems to be the cleanest iteration so far with a functional purpose and still matches the chrome finish (at least good enough for my own pleasure).

Another compelling reason for adding an active group heater is that it massively reduces the warm up duration.

To put the warm up process into a graph, I attached a bare temp probe onto the group bell and log it using Artisan:

The results

Note: I think the wavy reading is just noise or due to how the probe is attached - a dip can show up even if I just walk by (presumably from the air draft)....

It took only about 30 minutes to get within 2 deg C of idle/equilibrium temperature! Without the heater, it would take almost 1-1.15 hour to do the same (not exactly a problem as it's always on a timer switch regardless, but this is super nice!).

I've also developed a habit of flushing some water before the first shot of the day (performed @ 47 & 48 min). It was just to get rid of the water sitting in HX and replenish it with fresh water - not sure if it really does anything but why not. Regardless, the group temperature reading was barely affected. Same with pulling a shot @ 53 min.

Photo: My DIY restrictor PF - just an unused PF with a 3/8" needle valve attached.

More photo:

Photo: Dual-PID : One for group heater, one for boiler (just so I can get rid of the pressurestat clicking noise). I am using the enclosure front plate as a lid to cover the display and it would be flipped down normally (the LED is too bright in the morning otherwise).

Naturally, of course I would be curious about the brew temp next (where it really matters)...

This is how I measured the temp (below)..seems to work most of the time if not for the minor leak...

Since I can control both the group and boiler temps independently, I can now do some fancy stuff like running it like a saturated group instead of the natural declining temperature of a classic lever group. :lol: Reasonably temp stable for back-to-back shot too (about 2 min+ apart).

Photo: Saturated-group-like temperature profile, with a shot from 2 min+ before laid in the background (the faded line). One note though for accuracy sake, although the temp was quite flat, but I don't think it was as flat/pointy as shown here. The Artisan still did some smoothing despite having disable all smoothing options.

I haven't spent enough time on optimizing the temperature (is there even one??) and to try all temperature combination. However, so far it's been a great feature. I was able to dial my temperature easily to accommodate some of the experimental light roasts I had, and still get some pretty great/drinkable shots out of them.

Closing thought
So far the group heater (new iteration) been at work for about 1+ month - it has worked so well while looking quite native, so I thought it's worthwhile to finally put it out there for sharing. With a bit of patience (waiting for items to arrive from oversea), the entire mod isn't exactly expensive (likely cost less than one of your custom-made tampers ;p).

At the moment, the cables are just tucked towards the side..I am/was considering drilling a hole in the face plate and using an armored cable to sheath the cables (that'd look nice...) but am still considering it as I mentioned I didn't want to make any permanent modification...

Some caveats that are worth noting:
(i) The band heater for my specified dimension ended up being about 60W @ 240V. I am not sure (they probably can) if it can be made to a similar/higher watt rating for 110V.
(ii) If your group is a dipper style, there is a limit to how low your boiler pressure can go without impacting the preinfusion pressure (the Aurora is HX-fed with line-pressure preinfusion so I've gone as far down as 0.1 bar boiler pressure to accommodate a higher group temp).
(iii) If your group runs hot naturally even at low boiler pressure, you will need to use a thicker/heat-insulating gasket for it to provide a wider working temperature range.

Anyway, that's it! Hopefully it's something new/interesting read for you and you get something useful out of it, especially if you have been considering doing the same (adding group heater without irreversibly modifying the machine).
★★ Quite Helpful

User avatar
Posts: 1724
Joined: 18 years ago

#2: Post by Balthazar_B »

Sam, this is a really cool mod. Since it's a dipper variant, I'm not sure it's a good fit for my Bosco -- or even that the Bosco needs temp modification, from what my taste buds report -- but it's a fascinating approach to tweaking a lever machine with a light touch. Thanks!
- John

LMWDP # 577

Posts: 2973
Joined: 10 years ago

#3: Post by OldNuc »

A very interesting and novel solution to a common issue with the older commercial lever machines. As long as the objective is a higher stable group temperature the basic machine design is of no actual significance as the boiler operating temperature determines the nominal minimum group temperature only. This is a good fix for increasing the average idle group temperature and allowing true control of the increased temperature without the requirement for fiddling with boiler pressure/temperature and magic flushes.

This should appeal to the light roasted coffee crowd immediately.

I would just bundle the wiring up in a metallic colored spiral cable wrap and tuck it away. Easy to do and low cost.

Posts: 550
Joined: 17 years ago

#4: Post by Paolo »

I am impressed Sam!

If there is one shortcoming with some of the older levers like the Aurora and the Lambro, it is the really long warmup time.
I like the way that it is all entirely reversible.

How long did it actually take you to set this up, once you had the parts?

User avatar
Supporter ❤
Posts: 1723
Joined: 13 years ago

#5: Post by pizzaman383 »

That is a really solid solution. I designed my double-dipper to solve the same basic group head temperature problem. I might like to try using using the strip heater instead of or in addition to my plate heater.

Where did you source the heater and the hose clamps?

I did some experimenting with different temperature profiles by changing the group head temperature, the water temperature, and most importantly the delta between them. At the time I did it I thought that the standard dipper temperature setup was best but honestly I didn't do enough testing that I am certain of the results.

I look forward seeing the results of your testing and what specific temperature settings you use.
LMWDP #551
“Taste every shot before adding milk!”

User avatar
Team HB
Posts: 13820
Joined: 19 years ago

#6: Post by another_jim »

Wow! I actually had to look a few times before I realized it was wrapped around the group bell. Great idea!
Jim Schulman

User avatar
Team HB
Posts: 14294
Joined: 14 years ago

#7: Post by drgary »


Thank you for sharing that ingenious mod. I will be interested to read of the taste differences you're describing between a typical lever pull, where group temperature drops during the shot versus where the temperature is steady. I have thought that the temperature drop helps avoid overextraction and makes a better shot, but I have not put that to the test as you have, and I typically don't drink extremely light roasts. I have long had my Conti Prestina commercial lever modded with a PID that controls boiler temperature. I have no group heater. The Prestina is a dipper machine with a small boiler (4.5l), and a large commercial group (weighing about 18 lbs.). Adjusting boiler temperature with the PID yields excellent control for dialing in a coffee and makes adjustment very easy. The Prestina heats pretty quickly for a commercial lever, about 40 minutes, and I have it on a timer.

What I WOULD do for a good cup of coffee!

User avatar
[creative nickname]
Posts: 1832
Joined: 11 years ago

#8: Post by [creative nickname] »

Super cool idea-thanks for sharing it!
LMWDP #435

samuellaw178 (original poster)
Supporter ♡
Posts: 2483
Joined: 13 years ago

#9: Post by samuellaw178 (original poster) »

Thanks all for the comments and ideas, appreciate all of them!
OldNuc wrote:This should appeal to the light roasted coffee crowd immediately.

I would just bundle the wiring up in a metallic colored spiral cable wrap and tuck it away. Easy to do and low cost.
Thanks for that Rich, I will look into that. Easy to do and low cost are exactly the philosophies I go by. :D
Paolo wrote:How long did it actually take you to set this up, once you had the parts?
Hi Paolo, after all the parts have been confirmed, the installation process took less than an hour. But the 'hidden' time is the many hours of fun spent playing with it after installation and plenty of hours before installation musing over different ideas and shopping around. :lol:
pizzaman383 wrote:Where did you source the heater and the hose clamps?
Both from Ebay! I get you about not experimenting much beyond the initial best setting - once you get a temperature combination that works, there is really little motivation to keep changing. I think I am still in the honeymoon phase with my new-found ability, so I've been changing it quite a bit to understand what impact it has in the cup. But once this settles, I think I might just stick to a few go-to/standard temp profile to suit different beans/roast levels.
drgary wrote: I will be interested to read of the taste differences you're describing between a typical lever pull, where group temperature drops during the shot versus where the temperature is steady.
Hi Gary, I can give that a try and report back. The main variables so far have been my home roasting qualities making apple-to-apple comparison a tad difficult (unless the Nano 7 is here :wink: ). Though even without the heater mod, I get only about 3-4 deg Celsius drop (compared to the Conti Prestina which I understood has a unique temp profile among levers).

I personally am very sensitive to brightness in coffee (yet I like to try and make light roasts work because the aromatics from light roasts can be so enticing and dreamy - if only I can get half that aromas translated into the cup without the sourness...). My impression so far is the flatter temp profile (or in some case, increasing) seems to handle them better/more palatable by managing the acidity. Before the mod I had tried flushing the group to get the brew temp higher, but for some reason the end results weren't the same and not as good (nor as repeatable - I do prefer to avoid using flushing as a way to manage brew temp).

Another quite noticeable difference with the flat temp profile is the recovery speed in between shots - the group head temperature is almost unfazed by pulling a shot or flushing as shown in the OP. This is because the incoming brew water is less hot, but the group stays hotter (hence with ability to cool down faster back to the equilibrium temp). Seems to add half a cookie point for consistency if that's the goal. But getting the traditional temperature creep (in quick successive pulls) is't a bad thing either as you can get to have different impressions of the same coffee at different temperatures.

Posts: 236
Joined: 7 years ago

#10: Post by Madman13 »

I used a vented screw to route the wire for my group thermocouple (stick on) in an unseenn way on my Londinium. Perhaps you could do the same (may need to find one with a bigger vent for those thick wires