Hay, good question.
Silvia @ 109, t.c. on top of boiler.
i know the "correct" pid temp is unique to every machine, thermostat placement and bean, but im interested to know what everyone has theirs typically set to? also include your machine/model etc...
Im running a gaggia classic set at 93/94 with the rtd in the original thermostat location.
Hay, good question.
Silvia @ 109, t.c. on top of boiler.
With the Classic Id try upping it to around the 101-102C sort of mark and then pull a few shots and fine-tune according to the results in the cup. Shots that are too sour means the brew water is too cool, bitter shots usually indicate the opposite..... everything else being equal that is ::).Originally Posted by early_morning_ link=1203038332/0#0 date=1203038332
With Javabeens PID now successfully residing in my silvia, my last few days have involved a lot of shots and fiddiling around with set points.
On my Silvia [RTD in thermowell] and with a 10 day Peru Villa GE roast [stopped right at the onset of FC], Jazz on the radio, CS on the computer, and the luxury of a cruisy Sat morning to enjoy the fruits of my labour, I found the following.....
SV=107 Temperature is a tad too hot and the mouthfeel seems a little thin [which may be related to the higher temperature], The shots arent bitter/burnt but taste a little metallic and seem to lack body.
SV=103 Too sour..................sink shot.
SV=105 First reaction to my front palate is: too sour, but there is underlying an sweet/chocolateynessness that quickly overrides the initial sour hit. Mouth feel and body is good and the taste lingers long with rich chocolate and a pleasant underlying subtle sourness.
Pretty happy with this, I like the complexity.
SV=105.5 Even better, The initial sourness is still present but only as a subtle hint through the crema and it would better be described as a fruityness [I cant quite pick it, but maybe orange]. The chocolate overtones are present through the whole shot and the body/mouthfeel is the biggest I ve ever experienced for "latin" SO.....Wow!
SV=106 No sourness, all chocolate, same great velvety body, great finish. This is probably the textbook "sweet-spot" and many would probably prefer this to my SV=105.5, but I confess to prefering the "brightness" and complexity Im getting at 105.5.
Im really surprised that I can pick out these temperature differences, but Id add that the +/- 0.5C change only seemed detectable between the 105-106 point, but otherwise first crack is certainly detectable to my pallate.
Im blown away by the consistancy the PID is giving me, having a great morning!
Interesting that you found 107 is too high.
My PID is a Jims PID Kits one for the Silvia and its set at 109.8 C. Thats the value it comes with. I havent tried playing with it as my shots seem OK.
There are many people here with a PID Kits Silvia - would be nice to know if they have found it needs changing.
It depends on a few things.
With thermocouples mounted on top of the boiler, there is apparently a 2-3deg difference depending on whether you mount under the brew thermostat or the steam thermostat....incredible when you think about it, since they are only 10-15mm apart.
Older Silvias have a thermowell, if you use this to mount an rtd [it screws into it] the "read" boiler temperature will apparently be 2deg lower.
Silvias boiler design has changed slightly over the years, this may also make a difference.
Jim PID kits also come with a special thermocouple that has been "properly" calibrated, I have no idea whether my rtd is calibrated or not.
Ultimately, I dont care, we are all measuring an arbitary temperature at the boiler, and what is really important is the water temp through the puck....Unless you have a Scase, there is now way of accurately measuring this, so ultimately you can only go by taste.
You should try different temps: up the SV a degree, try it, go down a couple of degrees and try it.....Let us know if you what you find.
It can only aid our pallate.
3 or 4 Wire RTDs are self compensating as far as lead resistance is concerned so that the only variation being considered is the actual potential across the calibrated resistor (RTD). 3-Wire RTDs are the most common and I would suspect that this is what your installation would be using. Heres a bit of a spiel about them here..... :)
I just measured the water out temp using the TC/meter from CS. I removed the basket but kept the group head in. When you put the TC of the meter up the spout you can just see the tip of it nicely centered in the group looking down from the top. Hence the water will hit there nicely. I ran it a few times to warm the TC probe.
The water temp indicated is 95 C at the outlet when the PID setpoint is stable (i.e.109.8). Then as the shot continues the output water temp drops to about 87. So the first water stream is at 95. If you collect the water in a foam container will only show 87 as that initial 95 is lost in the larger quantity of lower temp water.
If you want to get a more accurate indication, you really need to do these measurements during normal brew flow rates, as thats how the system has been designed to work. If you position the t/c bead on top of a coffee puck and then lock into the group, this will give you the most meaningful data. The t/c cable gets a little beat-up where it has to negotiate the PF/Gasket interface but survives the exercise ok. Ive been using the same t/c for this sort measurement quite a few times and the cable insulation is still quite viable :).
Thanks for that link,
FTR... I have a 2wire RTD, therefore, given its design cannot compensate for lead resistance, the Vout value seen by the PID will be slightly lower than the actual RTD output.
Hence another reason why my setpoint may seem lower than others.
Nice experimentation, as Mal says you really need a puck to emulate the flow rate, but it does show that the water temp hitting the puck is 95 deg which is right in the zone.
Back to the original question,
with my current setup using a type T TC, mounted under brew thermostat,
I have found that best PID temp setting is around 104 deg C.
This was checked using the method suggested by Mal above,
and found to give brew temp around 94 to 95 from memory.
(Its not a PID kits setup)
If I set my PID to 108-109 brew temp is too hot,
but from 104 I can go up or down a degree or so if required,
to allow for different beans types and roast levels.