Controlling temperature in the flask. SPD

Hi. I’m new at distillation and on this forum.
According to information given here Short Path Distillation Procedure | Summit Research I have to slowly raise the temperature in the flask. But every time I set 40° on my controller it passes it and goes up to 120° very quickly. What am I doing wrong? What is the right way to raise it slowly?
P.S English is not my language. Apologies for any mistakes.

I use this temperature controller to apply mild heat to small areas of the boiling flask when starting the run because most of the time my stir bar is stuck in cold distillate at the bottom of the boiling flask. Once everything is warm and spinning around 120°c and under 75%-full vac I turn the mantle up the to amount of ml I have in the boiling flask. If I were running 2L of material I would start the BF heat at 10-100ml

The sensor used to monitor the boiling flask temp in your case sounds like it is placed into the flask itself and not outside the flask and monitoring the heating mantle directly. I have seen summit SOP for years now online and there are a number of problems with some of the info but if you monitor the oil for temperature instead of the heating mantle itself as is a typica summit SOP then you are bypassing PID heat control effectively at lower temps especially but certainly will see wild overshooting of temps like you do.

Irregardless of who makes the unit, placing the PID control into the oil to be heated and not as a monitor and precision measurment of the heating element itself outside the flask and between the glass and the mantle is a piece of design criteria I consider a flaw for two reasons at least. I have seen Summits older pages that talk about running a PID controller manually all over the place as things get going and this is indicative of a misunderstanding of the purpose of a PID control in my opinion. So what you get left with are vague references to “raising it slowly” when in fact a correct placement of the PID control sensor would raise the temp to precisely just where you want it. I realize many folks might run this way but I am sure of what I speak when I point out you are seeing a classic symptom of a design flaw insofar as having the sensor put into the oil and not outside the flask as I outlined.


This makes a bunch a sense. Id imagine you would still want to have the temperature of the BF with an external thermometer for data references and the head temp as well. I thought the oil at some point would be just about the same as the mantle, but i guess that wouldnt be correct until a certain temperature which would probably vary from setup to setup based on a number of parameters.

Thanks for this insight. Already rethinking my setup


You can get a better controller with both power regulation and temp control than the J-Kem’s and Glas-Col’s, and for way less money. The power regulation of the Auberin’s is minutely adjustable, 1-99%, and you can choose controller set up options such as to hold a temperature or Limit it (full stop) depending on your needs.

My write up so far…


Thank you for replying. What kind of control sensor do you have? There is no way I can place mine between mantel and flask.

1 Like

My sensor is a twisted wire pair that constitutes a thermocouple. Thermocouples are simply two wires twisted together. They often have metal barrels or sleeves they come in but on mine I clipped that off and just twisted the wires together.

This way I just lay it under the flask tucked into the fiberglass mantle a bit and the flask lays on the wire just fine.


Very creative. Thanks. I’ll try that.

Right on Beaker!!! Get the sensor in contact with the bottom of the boiling flask.

Btw, I checked out one of these super cheap K thermocouple thermometers (TM-902C,) works ok just like they’ve been saying at YouTube. Get several extra open ended sensors to go with it ($.99 or less each,) and you can play around with sensor placement before connecting your controller up.


it says -50C, k-type should be able to read lower than that.

have you tried it at -80C?

anyone else got a recommendation for appropriate thermometers at that temp?
everything I’ve run into seems to give up below about -50C.

have you tryed a type-T ?

you need an extension wire to read the from 0c on up.

Omega has a plethora of references and products, and will sell to anyone in any quantity, but their products aren’t inexpensive, and many are available only on a backorder basis (you order and wait…)

K thermocouple table in F:

Thermocouple types:

yeah, I should spend some poking around on Omega’s site again. It’s been quite some time, and I told my team I’d do it days ago…you’re right, they’re not inexpensive.

I’ve seen that a k-type should give usable signal at lower temps that -50C, but my harbor freight multi-meter stops registering below -50 with a k-type, and the instrument pictured above also seems to suffer from the same limitation.