@ Jeremy:
I ordered one of those OmniTesters made by smok that Andria graciously tested for us on the sigelei. I did some further testing with it and it seems to be pretty accurate. If I set my V3 to 4.1 volts, it says 4.1 volts on the meter. It "rounds up" or "rounds down" to the nearest 0.5 volt, but for the most part it's pretty decent.
At any rate, you mentioned testing a 1.3 ohm coil on one of these OmniTesters to see what the Sigelei could put out. I used about 1.5 inches of twisted kanthal that I had laying around on my KayFun deck (not in a coil, just a circular shape between the screws) and ensured it read 1.3 ohms on the Sigelei. Occasionally, I would switch the display back to "resistance" to verify that the 1.3 ohms was still correct. I starting increasing the voltage while I was in voltage mode and I got up to 4.3 volts on the V3 while the Omnitester mirrored the same voltage I had set. At 4.4 volts, the OmniTester still showed 4.3 volts, and any higher setting of voltage on the V3 did not increase past 4.3 volts on the Omnitester.
Before I continue, I need to explain that I opened up my OmniTester and connected a positive and negative wire that pass through a hole in the OmniTester enclosure to the outside so I can attach another meter. I connected my "graphical display Fluke meter (867B)" and I was able to confirm that when the V3 was set at 4.3 volts, the PWM waveform was at 98.6% duty cycle. At 4.4 volts on the V3, the graphical display showed the waveform completely flat at 100% duty cycle, and the Omnitester still showed 4.3 volts. My graphical meter displayed 4.37 volts, which should have been pretty close since the waveform was at 100% duty cycle (straight DC).
During this test I was using a single 18650 battery (Efest IMR 2250mAh V2).
1.3 ohms at 4.3 volts = 3.3 amps. I'm not sure how, but I was able to exceed the 2.5 amp limit with the boost circuit when using a single battery. Your guess is as good as mine regarding why I was able to do that. I should probably do some more testing with other resistances and see what I come up with.
Another significant thing I wanted to share...
If you recall, I stated in an earlier post that I was putting together a spreadsheet calculator in Excel using the formulas for PWM and the resulting output voltage and power. I also mentioned in that post that if I used my digital meter to determine the duty cycle at a specific setting on my V3, the duty cycle calculations from my spreadsheet were inconsistent. Now I think I finally understand why that is and what is really going on.
The peak pulse voltage of the PWM signal coming from the V3 is supposed to be 6 volts. However, I noticed that as soon as you introduce a resistance, that 6 volts gets loaded down and drops to some lower voltage. The value of the voltage it drops down to depends on the atomizer resistance and the voltage setting. All my calculations were using a steady 6 volts as a baseline PWM peak voltage, and that's why the duty cycle measurements I took with my meter didn't match my spreadsheet calculations for a particular voltage or power setting. I assumed that the 6 volt pulse voltage would be regulated by the device when a load was introduced in the circuit. This doesn't appear to be the case.
I found that instead of regulating the output pulse voltage, the device simply increases the percent of duty cycle to compensate for the voltage drop. The MCU monitors the output and increases the duty cycle until the output voltage once again matches the setting in the menu. It's actually a pretty smart strategy. They were able to fabricate a circuit that provides a relatively accurate output without over-complicating the design (fewer components/less space required/reduced cost).
If you go back to the above example where I found that the highest voltage I could get was 4.3 volts at 100% duty cycle, it validates this concept. If the assumption is that the output peak voltage in the PWM waveform is always 6 volts, shouldn't my reading be 6 volts instead of 4.3 volts? At 100% duty cycle, a PWM pulse train appears as a flat DC voltage, so anytime you have 100% duty cycle with a 6 volt peak pulse voltage your RMS output should read 6 volts. However that isn't the case. Since the peak pulse voltage has now dropped to 4.3 volts under load, the output also reads 4.3 volts when it's at 100% duty cycle. Makes sense, right? That being said, it should also be understood that the maximum power this device can deliver is also going to be determined (at least in part) by the maximum duty cycle of the PWM output when a specific load and voltage setting are applied.
I was eventually able to include a calculation in my spreadsheet that corrects for the voltage drop and determines what the actual peak pulse voltage should be. However, you must measure the percent of duty cycle (with a meter that is capable of doing it) and input that value into the spreadsheet. The resulting "corrected" value of the pulse voltage can then be input into the appropriate cell in the spreadsheet to overwrite the "baseline" 6 volt pulse value used for all the output calculations. Once that's done, the spreadsheet recalculates everything at the corrected peak pulse voltage. That let's you validate that the output will produce the same voltage or power you were originally simulating at the 6 volts peak pulse baseline, but at the corrected peak voltage. I'm still playing with the spreadsheet to see if I can automate some of that... I'm hoping that I'll soon have an easy to use spreadsheet that I can share with anyone who's interested. With my luck, I'll finally get it completed only to discover that someone else made one 2 years ago. Doh!
I'm finding out that not only is the spreadsheet useful and a great learning tool in and of itself, but also that the above pulse voltage correction calculation is useful when you are trying to figure out what your output should be when you don't have any other way to accurately measure it (ex. you have a meter that can't read RMS or you want to verify the voltage setting on your device). Again, the caveat is that you must own a meter that can read % duty cycle (many digital meters appear to do that these days). If your meter reads DC (most do), you should be able to accurately measure your output voltage so long as you can get it to reach 100% duty cycle before it hits some other limitation. Directly measuring your duty cycle with a capable meter will obviously also verify that for you.
Sorry this was so long winded!
I ordered one of those OmniTesters made by smok that Andria graciously tested for us on the sigelei. I did some further testing with it and it seems to be pretty accurate. If I set my V3 to 4.1 volts, it says 4.1 volts on the meter. It "rounds up" or "rounds down" to the nearest 0.5 volt, but for the most part it's pretty decent.
At any rate, you mentioned testing a 1.3 ohm coil on one of these OmniTesters to see what the Sigelei could put out. I used about 1.5 inches of twisted kanthal that I had laying around on my KayFun deck (not in a coil, just a circular shape between the screws) and ensured it read 1.3 ohms on the Sigelei. Occasionally, I would switch the display back to "resistance" to verify that the 1.3 ohms was still correct. I starting increasing the voltage while I was in voltage mode and I got up to 4.3 volts on the V3 while the Omnitester mirrored the same voltage I had set. At 4.4 volts, the OmniTester still showed 4.3 volts, and any higher setting of voltage on the V3 did not increase past 4.3 volts on the Omnitester.
Before I continue, I need to explain that I opened up my OmniTester and connected a positive and negative wire that pass through a hole in the OmniTester enclosure to the outside so I can attach another meter. I connected my "graphical display Fluke meter (867B)" and I was able to confirm that when the V3 was set at 4.3 volts, the PWM waveform was at 98.6% duty cycle. At 4.4 volts on the V3, the graphical display showed the waveform completely flat at 100% duty cycle, and the Omnitester still showed 4.3 volts. My graphical meter displayed 4.37 volts, which should have been pretty close since the waveform was at 100% duty cycle (straight DC).
During this test I was using a single 18650 battery (Efest IMR 2250mAh V2).
1.3 ohms at 4.3 volts = 3.3 amps. I'm not sure how, but I was able to exceed the 2.5 amp limit with the boost circuit when using a single battery. Your guess is as good as mine regarding why I was able to do that. I should probably do some more testing with other resistances and see what I come up with.
Another significant thing I wanted to share...
If you recall, I stated in an earlier post that I was putting together a spreadsheet calculator in Excel using the formulas for PWM and the resulting output voltage and power. I also mentioned in that post that if I used my digital meter to determine the duty cycle at a specific setting on my V3, the duty cycle calculations from my spreadsheet were inconsistent. Now I think I finally understand why that is and what is really going on.
The peak pulse voltage of the PWM signal coming from the V3 is supposed to be 6 volts. However, I noticed that as soon as you introduce a resistance, that 6 volts gets loaded down and drops to some lower voltage. The value of the voltage it drops down to depends on the atomizer resistance and the voltage setting. All my calculations were using a steady 6 volts as a baseline PWM peak voltage, and that's why the duty cycle measurements I took with my meter didn't match my spreadsheet calculations for a particular voltage or power setting. I assumed that the 6 volt pulse voltage would be regulated by the device when a load was introduced in the circuit. This doesn't appear to be the case.
I found that instead of regulating the output pulse voltage, the device simply increases the percent of duty cycle to compensate for the voltage drop. The MCU monitors the output and increases the duty cycle until the output voltage once again matches the setting in the menu. It's actually a pretty smart strategy. They were able to fabricate a circuit that provides a relatively accurate output without over-complicating the design (fewer components/less space required/reduced cost).
If you go back to the above example where I found that the highest voltage I could get was 4.3 volts at 100% duty cycle, it validates this concept. If the assumption is that the output peak voltage in the PWM waveform is always 6 volts, shouldn't my reading be 6 volts instead of 4.3 volts? At 100% duty cycle, a PWM pulse train appears as a flat DC voltage, so anytime you have 100% duty cycle with a 6 volt peak pulse voltage your RMS output should read 6 volts. However that isn't the case. Since the peak pulse voltage has now dropped to 4.3 volts under load, the output also reads 4.3 volts when it's at 100% duty cycle. Makes sense, right? That being said, it should also be understood that the maximum power this device can deliver is also going to be determined (at least in part) by the maximum duty cycle of the PWM output when a specific load and voltage setting are applied.
I was eventually able to include a calculation in my spreadsheet that corrects for the voltage drop and determines what the actual peak pulse voltage should be. However, you must measure the percent of duty cycle (with a meter that is capable of doing it) and input that value into the spreadsheet. The resulting "corrected" value of the pulse voltage can then be input into the appropriate cell in the spreadsheet to overwrite the "baseline" 6 volt pulse value used for all the output calculations. Once that's done, the spreadsheet recalculates everything at the corrected peak pulse voltage. That let's you validate that the output will produce the same voltage or power you were originally simulating at the 6 volts peak pulse baseline, but at the corrected peak voltage. I'm still playing with the spreadsheet to see if I can automate some of that... I'm hoping that I'll soon have an easy to use spreadsheet that I can share with anyone who's interested. With my luck, I'll finally get it completed only to discover that someone else made one 2 years ago. Doh!
I'm finding out that not only is the spreadsheet useful and a great learning tool in and of itself, but also that the above pulse voltage correction calculation is useful when you are trying to figure out what your output should be when you don't have any other way to accurately measure it (ex. you have a meter that can't read RMS or you want to verify the voltage setting on your device). Again, the caveat is that you must own a meter that can read % duty cycle (many digital meters appear to do that these days). If your meter reads DC (most do), you should be able to accurately measure your output voltage so long as you can get it to reach 100% duty cycle before it hits some other limitation. Directly measuring your duty cycle with a capable meter will obviously also verify that for you.
Sorry this was so long winded!
