"ABC hits harder than XYZ set at the same voltage..."

[Rader2146]

I see this mentioned quite often in reviews and various threads and wanted to generate some discussion on the topic. The general consensus, with evidence from oscilloscope readings, is that these APV’s produce an unfiltered, pulse width modulated, output. I’m working on a post that explains pulse width modulation and the difference between filtered and unfiltered outputs in detail, but in general: Pulse Width Modulation (PWM) is using high voltage for a portion of the time and low, or no, voltage for a portion of time to produce an average voltage equal to the set desired voltage. Example: If you set your APV to 3.7v, the PWM profile could be 7.4v for half of the time and 0v for the other half (50% duty cycle), averaging out to 3.7v.


So why does this seem to make a device “hit” harder that another device set to the same voltage?
Here’s my theory, and it’s fully open to criticism and corrections as needed.


I am not going to type out all the formulas, so here are some references:
Ohm's Law Wheel
Ohm's Law Calculator

“Watts” is the name of the game. For sake of explanation we’ll use an average voltage of 3.7 and a 1.7 ohm atty/carto. This should produce an output of 8 watts based on Ohm’s Law.


Let’s look at our high voltage of 7.4v. With our 1.7 ohm resistance, Ohm’s Law tells us that we will have an output of 32 watts during the “high voltage” time period.


Low voltage period is easy…0 volts, 0 watts.


Now given the example of a 50% duty cycle, even though our average voltage is 3.7v, our average wattage is 16w!



I also believe that the thermodynamics have a lot to do with it. Think of an electric range. If you set the knob on the range to high, how long does it take to go from resting temperature to full heat? Now turn the knob to off. How long does it take to return to resting temperature? Comparatively much longer than it took to heat up. So when thinking about our atty’s and carto’s, we are rapidly heating them with 32 watts, but when we take the power away they are not cooling at the same rate as when they were heated. This results in a higher average temperature than if it were a steady heat at 8 watts.

Open for thoughts...

 

[Wishdog]

As an engineer, I'll have to point out that the devices might not be properly stating their output.

Also, the temp difference thing. The temp of the atomizer will be so much higher than the area around it for the few seconds it's on that the slight increase in the surroundings will not effect it. It doesn't take longer to boil water in a room at 66 degrees than a 80 degree room. Same principle.

Fanboyism plays a large role as well.

 

[Rader2146]

I choose not to be device specific as I know that accurate calibration is probably not very high on the Chinese manufacturers' priority list. However from the devices that I have seen tested the error does not seem large enough to make up the spread. Many times the Provari is used as the "gold standard" with statements of "XYZ at 3.2v is about the same as my Provari at 4.8v.". That is the origin of my post.

With the temperature, after re-reading, my post is a little misleading. I am referring to the on/off cycling of the regulator in measure of kHz. Not the not the fire button in measure of seconds. Thousands of times per second there is a rapid heating followed by a period of slower cooling. Resulting in an average temp greater than that of a steady 8 watts.

That's just my theory, I could be way off. My understanding of electricity if far greater than my understanding of thermodynamics. But I love a good discussion on either.

 

[silversx]

if that's the way pwm works then how does pwm devices last longer on the same batteries then none pwm.. if at 50% duty cycle it averages 3.7volts then shouldn't it be using the same amount of power as a non pwm at a steady 3.7?

i do think the higher peak voltage is what cooks the atty and gives it the hard hit.. the whole average voltage thing just doesn't make any sense.. you're heating your atty at double your set voltage with intervals of 0 voltage but the coil does not cool instantly between pulses.. it's almost like firing at 7.4volts for half the time you would fire at 3.7.. still going to be a hard hit no matter what..

kind of makes sense when people say the vmax at 3.0 hits like a provari at 4.5.. rough estimate vmax at 3.0 is firing pwm of 6.0 volts mixed in with 0 volts intervals.. but the coil can't cool fast enough during the 0 volt pulses but cooks like the sun during the 6 volt pulses.. however the 0 volt pulses does give some cooling effect versus a constant 6.0 volt non-pwm.. therefore you end up some where similar to a 4.5 volt non-pwm device like a provari..

plausible

 

[DC2]

I don't understand most of this discussion, but I like it.
In the end, maybe we will get a better understanding of the underlying physics.

And I do like me some physics.

 

[AttyPops]

Well,

1) I guess I'd like to see the oscilloscope readings and know what the battery level was at a given average output voltage.
because....
2) I read somewhere once that they used a lookup table to set the PWM duty cycle (there are probably as many methods as manufacturers). So they may set it according to effective average wattage or some such. It's just that everyone plugs voltmeters on em. lol. Which makes this very interesting.
3) IDK for sure but a lot of the PWM modulated devices seem to be on single 3.7 volt cells. True?
4) The 7.4v example would be better served with a switching regulator example. I wonder if it does that same thing with the wattage math. IDK maybe we could have the same discussion with those too.

Summary:
This is all evidence of the voltage squared part of the math... and why higher voltage is more efficient in most cases. But... That's not necessarily how the device is calibrated.... it may be calibrated to effective watts. And it usually isn't a simple PWM on 7.4 volt devices, usually used on 3.7v cells.

Just my thoughts at the moment. Watching with interest to see how it shakes out.

 

[VapingTurtle]

...
i do think the higher peak voltage is what cooks the atty and gives it the hard hit.. ...

kind of makes sense when people say the vmax at 3.0 hits like a provari at 4.5.. rough estimate vmax at 3.0 is firing pwm of 6.0 volts mixed in with 0 volts intervals.. but the coil can't cool fast enough during the 0 volt pulses but cooks like the sun during the 6 volt pulses.. however the 0 volt pulses does give some cooling effect versus a constant 6.0 volt non-pwm.. therefore you end up some where similar to a 4.5 volt non-pwm device like a provari..

Couple of problems here...

1) a heating element does not heat up that very fast (relative to the 70-150 bursts per second in PWM implementations); take a gander at your coil while dry burning... even when clean and hot it takes a half a second or so before it really glows.

2) The ProVari also uses PWM.

 

[thinkingaboutit]

As an engineer, I'll have to point out that the devices might not be properly stating their output.

I'd suspect this is key.

I went from VV PV a to VV PV b today with the same tank back and forth. They were quite different at the same stated voltage. Both batteries read at 4.0...but there is room for error there also.

 

[lannister]

Same setup exactly on different pv's? And different vapes? This is the intangible that most engineers deny. Must be the gods. Seriously, though, what is it?

I'd suspect this is key.

I went from VV PV a to VV PV b today with the same tank back and forth. They were quite different at the same stated voltage. Both batteries read at 4.0...but there is room for error there also.

 

[Rader2146]

Technically all switching regulators, buck or boost, use PWM. The difference is whether it is filtered by an inductor and capacitor that are properly sized for the application. This is why I stated in my first post " is that these APV’s produce an unfiltered, pulse width modulated, output" For boost regs an inductor and capacitor are needed in order to boost the voltage. A buck regulator doesn't need an inductor or capacitor, but most DC to DC applications don't like square wave.

I certainly could have missed them, but I haven't seen too many statements of hard hitting single cell boost mods. I assume this is due to the inductor and capacitor that are required for boost operation. Even if poorly matched they will at least provide some filtering. Most of the statements that I've seen have come from 2 cell buck regulators. Mainly VMax and Ovale V8 as I was following those threads regularly.

 

[Wishdog]

Same setup exactly on different pv's? And different vapes? This is the intangible that most engineers deny. Must be the gods. Seriously, though, what is it?

Don't worry what it is, just be happy it is.

Anecdote:
When I was young, I would sometimes go with my dad to the podunk airport in the middle of nowhere where he owned two planes he rented out. One was N704YP and the other was N66118. YP, or Yankee Papa, was a standard, meh Cessna 150. Just like all the others there. However, the small plane gods clearly smiled down upon the factory when N66118 was made, because it was clearly much better than the others. Even the same year, same model, same everything. But it felt more solid and flew better...even though it was the same exact shape with the same engine, you get the point.

The PV that wasn't doing as well...have you tried turning it off and then back on again.

 

[zoiDman]

Here is a Non-Technical look at Constant Voltage verses PWM.

If my Regulated PV is outputting say 3.7 volts when I Smash Down the Power Button , it sends 3.7 volts thru the Coil Wire and by some Voodoo Magic (or something that Can be Explained with one of Maxwell’s Equations – one with a Funny Upside Down Triangle Differential Operator type thing) the coil gets hot. And when I let my Sweaty Finger off the power Button, the Coil Wire Cools Down.

The Coil wire doesn’t Heat Up to the Given Temperature Instantly when the Power Button is pressed. It ramps up over a Discreet Time Period. Then it Maintains a Constant Temperature (if e-Liquid wicking is sufficient). And Finally, when the Power Button is Released, it cools down to a Ambient Temperature. And this Cooling takes a Discreet amount of time also.

Neat. So what is the Discreet amount of time it takes to Heat the Coil Wire and to Cool the Coil wire. I dunno. I’m sure there are Off the Shelf Differential Equations for stuff like this. But EVERYONE hates Dif-EQ’s so let’s just say it is 1/4 of a Second for fun.

Now what happens with the PWM? Well it is On then Off then On then Off then On then Off then On then Off then On…. Then Off.

But a Key Question is Are the On’s the same Length as the Off’s? And if they are, How Long are they?

If the On’s are the Same Time Length as the Off’s, Then we can do that Divide by 2 thing on the On Voltage and say the “Average Voltage” is ½ the On Voltage.

But is the Heat coming off the Coil with PWM the same as the as a Regulated PV? Most people here say No, It’s Hotter.

So I’ll make the Jump that the Time It Takes to Heat the Coil Wire Is Less than the Time it Takes Cool the Coil Wire. So With PWM, There is a Stair Step Heating Process to go from Ambient to Constant Temperature when the Power Button is pushed. Then the Coil is Hotter than “Average Voltage” because It is Going from On Voltage (Hot) to Off Voltage (Ambient) but the Hot wins over the Ambient because it is Faster to get Hot than it is to get Ambient (Slower). Then is falls to Ambient when Power is Released.

Not Exactly a Rigorous and Robust Scientific Proof but it is the Best I can do without Drinking.

Maybe someone who Does Like Dif-Eq and Thermo-D can chime in on Heating and Cooling Rates of an Coil Wire. Because that is what I think is going on with PWM

Delta Heating/Unit Time > Delta Cooling/Unit Time

 

[zoiDman]

... It doesn't take longer to boil water in a room at 66 degrees than a 80 degree room. Same principle.

Fanboyism plays a large role as well.

It doesn't take Longer to Raise the Same Volume of Water at 66, with the same amount of Energy, from 66 to 212 as it does the same amount of Water at 88 from 88 to 212?

I didn't know that.

By the way, if the Room was hotter than the temperature of the Stove Flame, would that Change this Closed System Puzzle?

 

[AttyPops]

Yeah, that's why I put in #4 above. But... the switching regs are as you say filtered, also they adjust the duty cycle (?via comparitors and such?) dynamically on the fly constantly. Not that I'm an expert. But I was trying to get to the question.....

Unless you can say what a device is doing (what duty cycle at what voltage) you can't really answer your question. You have no idea if it is really doing a 50% duty cycle if it is set to 3.7 volts. Like post #2 said... its reported voltage may not be true average voltage. That's why I would want to see the scope detail AND a measurement of the battery voltage at that point. They may be doing average wattage. Basically they can come up with any algorithm. And it probably varies by device.

So I don't know if the question can be answered, unless it's for a specific make/model and we get accurate readings from a scope and the battery input voltage.

I retract the "PWM = 3.7 usually" idea. That's why I put a question mark on it. Note that I did not mean switching regs when I said it though. (Hence #4)

 

[Wishdog]

It doesn't take Longer to Raise the Same Volume of Water at 66, with the same amount of Energy, from 66 to 212 as it does the same amount of Water at 88 from 88 to 212?

Not measurably, no. The loose definition of "boiling" takes care of that. And if we define that as when the water stops rising in temperature (rather than just getting to 212 which it might not do, and might reach slightly higher depending on ambient factors) thus introducing a period of time that is not instantaneous...

Taken the exact amounts of water at exactly the same temperature in the exact same pot which starts at the same temperature in both trials, duplicating the ambient temperature and pressure and making sure no impurities exist in the air or water, and starting the heating element at the exact same temperature, then no, there is no difference...because you couldn't do all that to save your life.

Edited to note my mistake in paragraph 2: See, I accidentally said same temperature of the air. Doh.

 

[zoiDman]

Yeah I’m not a PWM Expert either.

But I am of the Opinion that it is more like 2/3 On and 1/3 Off. Or perhaps Less It would make sense if the On’s were longer than the Off’s because that would take Lower Voltage to achieve a given “Average Voltage” than ½ On and ½ Off.

But I could be Way off. And the PWN is Probably not a Linear Function (Nothing ever is once you get out of Book Examples in some University Textbook) and Probably Changes as Battery/Batteries Voltage Falls.

 

[zoiDman]

Not measurably, no. The loose definition of "boiling" takes care of that. And if we define that as when the water stops rising in temperature (rather than just getting to 212 which it might not do, and might reach slightly higher depending on ambient factors) thus introducing a period of time that is not instantaneous...

Taken the exact amounts of water at exactly the same temperature in the exact same pot which starts at the same temperature in both trials, duplicating the ambient temperature and pressure and making sure no impurities exist in the air or water, and starting the heating element at the exact same temperature, then no, there is no difference...because you couldn't do all that to save your life.

So the Ambient Temperature in the Room Doesn’t Matter.

So it sounds like if we start with say 1000ml of Water at 66F and Boil it inside a Sub-Zero Walk-In Freezer it will take the same amount as 1000ml of Water at 66F if we Boil it inside a Pottery Kiln Cranked up to 2200F.

 

[DC2]

Not Exactly a Rigorous and Robust Scientific Proof but it is the Best I can do without Drinking.

Best... post... ever...

And if I was drinking, I would have liked it even more.


I'm not saying I agree or disagree, but that's neither here nor there.
And although I was a physics minor in college, I thought Maxwell was a ..........

And differential equations? Yikes!!
I got a solid B grade in that class, and I never knew what the hell I was doing.

 

[Wishdog]

So it sounds like if we start with say 1000ml of Water at 66F and Boil it inside a Sub-Zero Walk-In Freezer it will take the same amount as 1000ml of Water at 66F if we Boil it inside a Pottery Kiln Cranked up to 2200F.

No, and for the exact reason you picked those temperatures, because they are significant. I picked the temperatures that I did because they matter to us, and we're familiar with what the feel like due to the range of temperatures in our houses and the weather outside, but they are far enough away from the boiling point, but close enough to each other that it doesn't matter.

(Edited to add: We're also talking about air at that temp, which has a crap heat capacity compared to the water. And, I don't recall off the top of my head, but isn't it one gram of water evaporated is 22 liters of water vapor? (that can't be right...) But that wouldn't start affecting the ambient temperature around the water until around 190.)

The kiln would boil the water quickly. Flash off the water if it was a small enough amount. And if it was enough that the water didn't flash off, then the temperature would drop drastically in the kiln due to the latent heat of vaporization eating the heat up during the phase change.

And you'd have to pull a serious vacuum to boil water below it's freeze point. It can be done. But I've never boiled water at less than room temperature. Which, when I say that, I don't mean STP...we kept the lab around 77 degrees because that was more comfortable to us (rather than 72).

Try it at home if you have doubts. But you can't, unless you get some serious high end equipment. And that's the whole point.

The reason that pi is 3.14 is because 3.14159 is less than 0.002 off. And in what situation is being 0.1% off going to matter. As if calculations were always reflected in reality.

(you're going to make me start pasting engineer vs mathematician jokes here in a while)

Edited again: You can't get your water into the kiln without some evaporating off so you don't have the same amount anymore. See, the real world keeps messing up our perfect theories.

 

[zoiDman]

...

(you're going to make me start pasting engineer vs mathematician jokes here in a while)

Please Don't.

The Engineers being Smart are Always Much Funnier. And the Engineers can Usualy Spell Better. LOL