Oh wow. 20 micron. I have no idea what I can do. I've never seen it below 160 on this unit. I'm running an Edwards 30 at full tilt so that isn't the issue. All joints seem properly sealed. Suggestions?
Its a long post but perhaps this can help.
An Edwards 30 has the same capabilities as my Edwards 28 really and it can easily pull a short path rig or sublimator down to one micron and a bit lower if installed correctly. This should be measured closely to the connection point of hose to glass and should be done with nothing at all in your rig. If the pump cannot pull down to one micron connected to an empty distillation rig then that is the place to fix. The problem of course is how to measure it.
In order to accurately see readings that low you must employ a pirani style sensor. Even the HVAC high dollar sensors lag considerably (HVAC would not ever need a Pirani based system) and in theory could measure down to 20 microns as a bottom limit but accuracy suffers at each end of an instruments limits. I run both styles at times and when down below a few hundred microns the HVAC electronic gauge is simply inaccurate. However below 1000 microns the Pirani sensor based units are extremely accurate down to one micron but they aren't any use above about 1000 microns.
The reason I see high vacuum levels of a few hundred microns when the pump at blank off pulls down to ¾ of one micron is pretty obvious when you run a hyper sensitive sensor. I see that too on my rig(s). First of all I can eliminate leaks because the "leak down" only leaks down to about 1000 microns but is slow and gradual. A leak in the vacuum system leaks off nearly all vacuum and is a sharp "leak down" slope. So I can eliminate leaks as a problem in seconds or identify it as quickly and fix.
If there is no leak then here is what is happening in my rig. The
terpenes and low boiling temp parts of the resin boil off in such large quantities continuously that the pressure stays high because new gas is evolving quickly. I have seen this countless times. Typically a few hundred microns at first. The trick is to keep temps low but hot enough to keep the boil going. I use the pirani sensor to tell me when the
terpenes have nearly finished coming off because the pressure is directly proportional to the quantity of
terpenes left in the boiling flask if the temp is held constant. The pressure on my own Edwards EM 28 blanks off at ¾ of one micron at the connection to the glass. When running an intial boil for terpene removal the pressure during boil generally pulls down to a few hundred microns and no deeper at first. I keep the temps low - a MUST in my operations for purity. Just hot enough to keep collecting distillate.
As the terps and stinky stuff collect in the receiving vessel slow but sure the pressure graph will slope ever deeper. It goes over say a half hour from a few hundred microns (50 ml boiling flask so small amounts) and pulls down deeper until generally in the unit I use for terp removal it will pull down to 10-15 microns as the collection flask fills with terpene. At that point I can squeeze a bit more out by bumping the temp ten or twenty degrees and can get it to pull down between 1-5 microns but that is more because my next step, cryogenic sublimation, is intolerant of even trace
terpenes. I even use swing pressure techniques alternately dumping vacuum a thousand microns or so then immediately pulling down again which pulls out stubborn traces during the swing back down to deep vacuum.
Once the pressure bottoms below 20 microns the catch flask will have collected about all the terps that started in the boiling flask. The key is to start at room temp and pull the vacuum. I pick 100C as my first target temp. Then I run the stuff at that temp and deal with all the bubbling and muck of a first run. Then I begin bumping the temp up 10 degrees or so at a time. Patience is key for me because it takes a while for the oil to heat up to mantle temp.