Many thanks to Thundercastle for his input into this report
Some pre-manufactured coils are inaccurately described as 'ceramic' simply because they incorporate cosmetic or other sundry ceramic parts. But here we refer to 'ceramic coils' owing to the hard porous ceramic wick material (commonly pressed into a tubular bead). This bead normally has a coiled wire heater semi-embedded into its inner diameter (see pic).
The most common pre-manufactured coils instead incorporate cotton, which is wrapped in one or more layers around a mesh or coiled wire heater. Cotton has effective capilliary action, which helps maintain ample supply of juice to replenish that lost from vaporisation.
However, cotton scorches at just over 200°C which is not far in excess of d-juice boiling point (pure PG 188°C @1atm), and potentially unpleasant dry hits will almost certainly result if any part of the wick runs dry.
Ceramic coil types use hard sintered ceramic wicks in the place of cotton, which are reported to be far more resistant to heat damage.
The common non-ceramic mesh/cotton coil design
Sintered ceramic wicks are commonly used in cartridges, providing impressive performance in such a tiny atomiser.
For larger tanks, ceramic coils are often simply scaled-up versions of cart atomisers, the Vaporesso EUC ceramic and GT CCELL2 are two examples of these.
But do they work as well in their larger size? In my experience the answer for these two (and almost certainly others of a similar design) is a definite no.
The showstopping problem is capilliary action in the wick, which appears woefully insufficient to feed the heating coil at the required rate to maintain adequate vapour
Not to scale. Left is a tiny ceramic atomiser from a Ccell TH2 cartridge, and right is a much larger partially dismantled 0.3Ω Vaporesso EUC ceramic. Note the amazing similarity of the designs
For these I used two coils, a Vaporesso GT Ccell 0.3Ω installed in an NRG tank, plus a Vapore EUC Ccell 0.3Ω in a Veco tank. Both have SS316L wire. Although these coils outwardly appear different, the cores are visually identical and performed indistinguishably in my tests.
The first sign of trouble came when attempting to set up temperature control. My usual basic setup for TC is the following
Fill with pure PG
Set mod to TC-SS, 20W limit, and say 220°C (any temp that's way above the expected boiling point of PG at 188°C)
Hold down the fire button (no toke) and monitor the calculated temperature on the mod - which will then indicates the approx boiling point of the juice (at least what the mod thinks the boiling point is)
If coil and wick are working correctly, the temperature of the coil should remain limited at a maximum of very near the boiling point of the juice. Therefore, providing the temperature is set above the juice's boiling point, it should never be reached irrespective of no-toke or toke rate. So the mod simply drives the user-set Wattage limit until the fire button is released.
This principle is shown on the Arctic Fox graph below using a Kylin M tank filled with pure PG, Grus 100 mod set to TC-SS, 220°C and 30W limit (a ~7 second toke was used for the tests). Green line is the Wattage, red is temperature in °C.
The Kylin M showing exactly how the graph should appear during a toke. The red line is coil temperature rising to and then limited at the PG boiling point, with extra power going into vapour production. The green lin is Wattage which remains at set max value throughout fire cycle in mod's attempt to reach user-set value of 220°C. Set temperature will only be reached if wick started to dry out (i.e. a form of dry hit protection)
However, when testing the ceramic coils under the exact same conditions, the temperature continued rising past the PG boiling point and quickly reached the set 220°C, which caused the mod to throttle back the Wattage asshown on the graph below.
The 0.3Ω GT coil shown (although both ceramic coil types performed identically). From the start of the toke, temperature quickly rose above the PG boiling point to the 220°C set value, where the mod responded by throttling back the Wattage from its 30W limit to around 15W
The graph shape is indicative of juice starvation after the first second or so of the toke. From this it appears there is insuficient juice in contact with the coil to hold it's temperature at the boiling point.
I tried Thundercastle's tip of speeding the flow of juice through the coil. This "Thundercastle Suck" involves creating a suction in the air passages by almost totally closing the tank's air intake vents and using significantly greater toking suction force to restore normal airflow. The hope was that the vaccuum on the inner part of the coil creates a greater flow of juice through the wick. Using this technique, a slight improvement resulted, but nowhere near enough to restore performance to normal levels throughout the entire fire cycle
Same test conditions as previously apart from with vents nearly closed but using much harder suction on the toking to restore normal airflow. Temperature still rose to 220°C but at a slower rate, and the Wattage throttling occured later and reduced at a more gradual pace.
Of course, I was keen to test out the cermic coils with dimitri juice. But the above tests were carried out with pure PG, which is expected to be considerably less viscous (and therefore flow faster through the wick) than a dimitri/PG even without any added VG. In fact Thundercastle reported that in his experience, weaker mixes of 1:6 performed better than the more ususal 1:4 or stronger, probably owing to this issue.
I decided to test the 0.3Ω EUC ceramic coil with 1:4 dimitri/PG mix (no VG) and a starting power of 25W. Upon vaping, problems were immediately apparent as follows.
Insufficient vapour produced, which seemed to become ever-weaker to near 100% air after a few seconds
Totally inadequate effect from the dimitri
Taking multiple successive tokes did not help, owing to time taken for the coil to recover from the previous toke
Determined hard toking to try and get some effect produced an odd-tasting vapour which left me with a slightly sore throat but no stronger dimitri effect
I tested the "Thundercastle Suck" (closed air vent/hard sucking) trick mentioned above but this did not seem to help much either. Owing to the terrible performance and total lack of confidence in these coils I decided to abandon the remainder of the planned tests.
Although I only tested two sintered ceramic coil models, the basic design is common to all true ceramic types I could find currrently on sale. Therefore IMO they are definitely best avoided.
Although smaller versions (as used in cartridges and smaller 'mouth to lung' pod systems) may be impressive, coils with ceramic wicks for larger tanks such as 'direct to lung' or 'restricted direct to lung" appear totally ineffectual for dimitri juice vaping.
There is strong evidence that capilliary action of the wick is much too slow to perform adequate vaporisation
Some techniques to achieve the best from them are as follows, none of these can disguise their extremely sub-par performance
The closed vent/hard suck technique (the "Thundercastle Suck" as decribed above)
Keeping the juice viscosity low by using weaker mixes (say 1:6) & with only minimal added VG
Using temperature control to auto-throttle the wattage (an unreliable solution bearing in mind how finnicky temp control can be on even the best devices)
Using a gentle slow 5 second toke with VPC mode to tailor the mod's power output to match the vapour-producing capability of the coil/mix. For example
0 to 1st second after fire button held - 20W
1st to 2nd second - 15W
2nd to 3rd - 13W
3rd to 4th - 12W
4th to 5th - 11W
from 5th second onwards - release fire button and wait at least 40 seconds before re-trying
