Zirconia balls instead of rubies in ball vapes?

[Radwin Bodnic]

I'm a bit confused by these numbers.

	Al2O3	SiC	ZrO2
Thermal Capacity in J/g-°C	1	0.7	0.4
Thermal Conductivity in W/m-K	25	125	2.7

By thermal capacity do you mean Specific Heat ? The S.I. unit is J/Kg.K (Same dimensions : energy/mass.temperature, but different units.)

 

[GoldenBud]

I'm a bit confused by these numbers.


By thermal capacity do you mean Specific Heat ? The S.I. unit is J/Kg.K (Same dimensions : energy/mass.temperature, but different units.)

yeah it's not the cp (Specific heat) values. SiC has a very low cp value, almost like glass. the other two have like 1000x more value to their specific heat values.

 

[SixStringToker]

I had ChatGPT churn out a simple table. The numbers that I double-checked were within the appropriate ranges, but... ChatGPT.

 

[Radwin Bodnic]

I had ChatGPT churn out a simple table. The numbers that I double-checked were within the appropriate ranges, but... ChatGPT.

That's useful ! Do you remember the temps at which those values were determined ?

Its clear that the Thermal Conductivity of Boro, Quartz and Zirconia makes them ideal for butane powered ball vapes. Being heated intermittently they need to retain the best the stored energy.

 

[SixStringToker]

Do you remember the temps at which those values were determined ?

I didn't check and ChatGPT didn't tell me.

I gave it a very brief prompt for that table. I was also sorta testing just how brief I could be:

Thermal properties table: Soda-Lime Glass, Aluminum Nitride (AlN), Ruby (Corundum), Alumina (Al2O3), Zirconia (ZrO2), Silicone Carbide (SiC), Titanium G2 (Ti), Silver, Copper, Gold, Quartz, Borosilicate. Columns: Thermal Conductivity (W/m-K), Specific Heat (J/kg·K). Sort Conductivity.

 

[fubar]

Air only doesn't triggers taste buds.You need particulates flowing through the air to taste something.

Yep. Exhaling through the nose is a good way to get a "taste" of what you just inhaled - but only the first time - the specific nasal receptors responding to vapour soon stop signalling from overload in my experience - my wife can smell the weed I've been vaporising for some time after I can no longer smell it from the overload I guess.

Taste buds are over-rated - there are only about 500 in your whole oral cavity and they "only" give 5 or 6 distinguishable "tastes" - sour, sweet,...umami from liquids passing through, so yes, only things dissolving in saliva from inhaled air are going to trigger taste buds and the vapour you inhale will not trigger tastebuds as much as swilling bong water around would do, but you will still get some sensation from whatever gets dissolved.

So important as they are, the majority of the complex terpene taste sensations must come from the back of the nose, where there are hundreds of thousands of differently tuned receptors and they respond to molecules in the air. This makes the whole topic of taste in vaporising so tricky and what we report as "taste" is probably coming from nasal molecule recognisers - and they are finicky and quickly/temporarily exhausted when stimulated strongly.

Full disclosure: I have tried inhaling vapour directly into my nose. Don't worry in case I ever offer you a bowl, because I do remove all snot from wand with ISO afterwards. Results are that I can "smell" the first part of the first nasal intake but the receptors all flake out really fast and the "smell" becomes hard to detect in the second and subsequent tokes. Interestingly, there must be some fast absorption in the nasal cavity because I do get a sense that nasal hits hit fast FWIW - anyone else care to experiment? Glassware designed for nasal use could soon become a thing!

 

[vapefritz]

By thermal capacity do you mean Specific Heat ? The S.I. unit is J/Kg.K (Same dimensions : energy/mass.temperature, but different units.)

Yes, specific heat capacity as listed in the tables I linked.

 

[SixStringToker]

Here's an updated table:

Differences from the last table are probably due to the fact that there's a range of values listed online for these things. Since it's a language-based AI model, the rest of the text in the prompt will essentially have an influence on which of those values it chooses.

 

[fubar]

While experimenting, swapping zirconium, ruby and titanium balls in a freight train over a couple of days, the thermal properties changed noticably in terms of PID overshoot and stability, meaning that the PID probably needed to be retuned every time the balls were swapped out - and I didn't realise until I started watching the PID more carefully.

PID tuning will change the extraction profile for any given ball material, so any real effect purely from different materials is confounded by changes to the PID profile, making it even harder to tease out the effects. I ended up with the zirconium bearings, and am currently retuning the PID to see if that changes anything. Got a tiodw mini to play with too if I really want some variables to sort out.

Anyone else noticed this and found noticable differences after the PID was re-tuned to suit the thermal profile of the new combination of materials? Searching for escaped 3mm balls on a dirty floor is exciting, but not quite enough fun to make me do it all again and retuning PIDs is best done from cold AFAIK to get the ramp-up phase right.

 

[SixStringToker]

PID tuning will change the extraction profile for any given ball material, so any real effect purely from different materials is confounded by changes to the PID profile, making it even harder to tease out the effects. I ended up with the zirconium bearings, and am currently retuning the PID to see if that changes anything. Got a tiodw mini to play with too if I really want some variables to sort out.

Anyone else noticed this and found noticable differences after the PID was re-tuned to suit the thermal profile of the new combination of materials?

The coil tuning isn't going to have much, if any, effect on the actual hit. There's just no way to get enough heat calories from the coil, thru the head and into the balls within the duration of a 15 to 20 second hit. Add to that the fact that you're actively cooling the balls with the airflow and likely have a *LOT* of momentum on that cooling before the pid controller starts ramping up the duty cycle. The controller has a major uphill battle at that point.

On the subject of duty cycle, think about how much your temperature drops during a hit. Maybe 15 to 20 degrees on most heads... 35 to 45 on the ones where the coil extends down past the ball chamber. That's likely not enough of an offset from the target/set temp on the PID controller for it to ramp up to 100%, so it may not drive the coil very hard to recover that lost heat; certainly not as hard as during the initial heat-up. Meanwhile, the balls have dropped 100F, 150F... maybe even more.

The closest I ever came to that level of efficiency was with one of my 16mm rTiTi heads. From hitting it, I actually felt like it was able to keep up with my hits - it felt like it had endless vapor. My temp logger setup (4 channel 100ms sampling meter, air pump, flow meter, tons of tubing, bit of blood) showed that not to be the case, however - the calories were definitely not being replenished at 1:1. That probably was my most efficient head, tho.

In the case of the current ball vapes, the coil tuning is about what happens between the hits, during the initial heat-up and the recovery. It can make an improvement in how quickly the coil reaches the target temp but keep in mind that, in most cases, the balls are NOWHERE near the same temp at that point; they lag behind by a lot, especially with the heads that have thicker ball chamber walls.

retuning PIDs is best done from cold AFAIK to get the ramp-up phase right.

That'll probably just drag out the process. I've never really dug into what the controllers are doing during the auto-tuning, but I bet it's a pretty simple process on most of them. Something along the lines of:

1. Shoot for XX degrees over the target temp and then XX degrees under the target temp. Calculate new PID values based on the overshoot/undershoot.
2. Using the new values from #1, repeat the process
3. Using the new values from #2, repeat the process
4. Done. Save the new values from #3

Edit: I just noticed the image with the table is broken. Here 'tis again:

 

[fubar]

the coil tuning is about what happens between the hits,

Good point - tuning is done while the head is just sitting heating up the planet without any air movement - so it probably won't have much effect on the 20-30 seconds when air is drawing heat out of the heat mass. Might just reset to the original FT/Auber settings that I wrote down - much less overshoot on startup and otherwise much the same performance...

 

[SixStringToker]

Good point - tuning is done while the head is just sitting heating up the planet without any air movement - so it probably won't have much effect on the 20-30 seconds when air is drawing heat out of the heat mass. Might just reset to the original FT/Auber settings that I wrote down - much less overshoot on startup and otherwise much the same performance...

I bet if the head were tuned such that the balls inside the ball chamber reached the target temp in the quickest possible time, we'd see a frighteningly MASSIVE overshoot on the coil temp. There's really no way around it with the current implementation of using the coil temp outside the ball chamber to control the ball temp inside the chamber.

I'd go with the auto-tune numbers and call it done.

 

[fubar]

using the coil temp outside the ball chamber to control the ball temp inside the chamber.

Being lazy, I'll just leave it - the stock FT settings and autotune are definitely different - autotuned coil ramps up fast and overshoots and even glows red for a while - might be optimising speed to steady state - doesn't do that with stock settings. No obvious differences in extraction profile in my hands.

Perhaps even more directly related to the extraction profile would be the temperature and volume of air about to flow through the bud, but not much help given long delays between pid control events and changes in heat being delivered to walls of a baller? Not that it is likely to make a lot of difference because to some extent, self-medicating humans have probably always adjusted their personal toking technique without much thought, to optimise (whatever that means) their medication, from whatever device they have to hand, somehow, for the tens of thousand of years we've been using the plant for fun.

 

[Gomaruana]

I've been saying this for awhile but Zirconia /ZrO2 balls Rock IN A Baller, I run all Zirconia in my tiodw and I'm running sic in my ceroma at the moment
I find the ceramic like ZrO2 and Sic while looking similar to Boro or ruby on paper they don't perform anything like those materials
They dump there heat very fast and they're slower to build up, I'm using them in a butane ball vape and they give a killer unique experience

In my coil units they dump the heat so fast you get this fantastic terpy rip right off the bat, I was so impressed with the Zirconia and sic performance I bought multiple kilos in order to get people to try them and give a good price, I really think they blow ruby out of the water without any competition

What material do you feel would work best for a wireless ballvape? I'm wondering how ruby / sic / zirc would compare for that usecase.

 

[SixStringToker]

Here's some data that I recorded over the weekend. I did each run twice from a fully cooled head and coil and got the same results each time:

That's with a temperature probe positioned in the middle/center of the ball chamber, surrounded by the balls.

The coil temp was taken with an external probe positioned on the coil close to where the coil's internal probe is. The external probe read maybe 10 to 15 degrees over the internal probe once the head was fully heat soaked. With the exception of the initial overshoot, that bouncing in the coil trace didn't show up at all on the display of the controller - once it recovered from the overshoot, it settled on the set temp and didn't vary. I think the external probe is just reacting more quickly to the duty-cycle than the internal one does.

It's interesting the the zirc actually climbs as quickly as the SIC when the coil is being driven at the full duty-cycle but quickly loses that momentum once the controller significantly cuts the duty cycle to correct for the overshoot and then maintain the set temp.

 

[Radwin Bodnic]

Here's some data that I recorded over the weekend. I did each run twice from a fully cooled head and coil and got the same results each time:

That's with a temperature probe positioned in the middle/center of the ball chamber, surrounded by the balls.

The coil temp was taken with an external probe positioned on the coil close to where the coil's internal probe is. The external probe read maybe 10 to 15 degrees over the internal probe once the head was fully heat soaked. With the exception of the initial overshoot, that bouncing in the coil trace didn't show up at all on the display of the controller - once it recovered from the overshoot, it settled on the set temp and didn't vary. I think the external probe is just reacting more quickly to the duty-cycle than the internal one does.

It's interesting the the zirc actually climbs as quickly as the SIC when the coil is being driven at the full duty-cycle but quickly loses that momentum once the controller significantly cuts the duty cycle to correct for the overshoot and then maintain the set temp.

That's some really good data, thank you @SixStringToker !
Interesting to note that ZrO2 has a bigger offset than Ruby and SiC. (Almost 25°F lower at stabilized temp.)

Could you do a cool down record ? It could be interesting to see how the inertia of the different materials differs.
Also, do you have boro and quartz balls on hand ? Andmaybe a vacuum pump ? (Okay, maybe I'm a little too enthusiastic…)

 

[SixStringToker]

That's some really good data, thank you @SixStringToker !
Interesting to note that ZrO2 has a bigger offset than Ruby and SiC. (Almost 25°F lower at stabilized temp.)

Yeah, I was surprised by that. I figured it'd settle around the same temperature but take longer to get there.

Could you do a cool down record ? It could be interesting to see how the inertia of the different materials differs.

I can add it to the list.

Also, do you have boro and quartz balls on hand ? Andmaybe a vacuum pump ? (Okay, maybe I'm a little too enthusiastic…)

I currently have 2, 2.5, 3, 3.5, and 4mm ruby, 3mm and 4mm zirc, and 3mm of sic, boro, soda lime, stainless steel, and quartz.

Re: the pump, my full test setup includes a pump hooked up to a rig, a flow meter, and 4 channel temperature data logger. I can log coil temp, ball chamber temp, herb bed temp, and... bowl wall temp maybe? It's not currently setup, but I'm hoping to get it back up and running this weekend. You can catch a glimpse of it on the coffee table behind the vape setup in this video:

Here's some data I recorded a year and a half (ish) ago at the bowl:

 

[Radwin Bodnic]

You can catch a glimpse of it on the coffee table behind the vape setup in this video:

Also a nice guitar and beautiful painting in the background.
I am definitely interested into quartz and boro performance. And in cool down data, now that almost everyone is going wireless...

 

[Cheebsy]

I guess that explains why some people say the flavour is better with zirconia balls, they're cooler! I don't understand why they're cooler though, does that offset remain at different temperatures?

Thanks for doing those measurements @SixStringToker

 

[SixStringToker]

does that offset remain at different temperatures?

I haven't looked, but I bet it does, though it may not scale linearly.

Also a nice guitar and beautiful painting in the background.

That painting is actually my Rabbit Air filter; it has different face plates.

It's an awesome air filter, btw. It's near silent 90% and it ramps up quickly within 15 to 20 seconds of me hitting a bowl in the same room.

 

[SixStringToker]

I guess that explains why some people say the flavour is better with zirconia balls, they're cooler! I don't understand why they're cooler though, does that offset remain at different temperatures?

Thanks for doing those measurements @SixStringToker

Maybe. I think the biggest difference comes from the roast profile. That's why I really want to see the temps at the herb-bed with the different materials and ball sizes... just need to motivate myself to get that setup running again.

With Zirconia, the temperature of the air hitting the herb bed probably ramps up more slowly so a larger time period of the hit will be while it's still ramping up through the lower / more flavorful temperatures. SIC, on the other hand, probably flies right through those lower temperatures and into the less flavorful zones. That same thing happens when we change ball sizes while keeping the material the same, too. I'm kinda expecting the 2mm or 2.5mm ruby to have a similar heating profile to the 3mm sic and 4mm ruby balls to have a similar profile to 3mm Zirc, but we shall see!