Breaktru Forum
eCigarette Forum => Modding => Topic started by: CraigHB on January 27, 2013, 04:03:36 AM
-
With the recent availability of the 3400mAh Pansonic NCR18650B, I'm thinking that kind of energy density is too good to pass up so I started looking at doing a booster mod with two of those cells in parallel. Unfortunately, the discharge curve on those is too steep for the set of electronics I'm using. I'd need to drop down to a supply voltage of 2.7V which is fine for the MCU, but gives me problems for the LCD and capacitive sensor. Those have a minimum 3V supply.
I was thinking there might be another high capacity 18650 cell with a taller curve I could use. In searching I came across this site;
http://lygte-info.dk/review/batteries2012/Common18650Summary%20UK.html
What a wealth of information. Check out the "Individual tests" link and the "Comparator" link. So browsing that site, I came across the LG 18650 D1 (http://lygte-info.dk/review/batteries2012/LG%2018650%20D1%203000mAh%20(Pink)%20UK.html)and the discharge curve really jumped out at me. Now, it's a new generation Li-Ion with a 4.35V terminal charging voltage, but that's not a problem. The charger controller I'm currently using is available in a 4.35V version. That would be the only thing I'd have to change hardware wise.
What's interesting is that because the cell operates at a higher voltage, it has similar energy capacity in comparison to the 3400mAh NCR18650B. Energy capacity is the integral of the discharge curve so you can eyeball it and see the curve between the two has similar area underneath. That can be seen easily using the comparator. You can also see that in the table data for the individual reviews. What that means is run time between the two will be similar even though the LG cell has less charge capacity.
I've only been able to find that LG cell for sale on eBay and I'm not sure it's the right one. It's a cell that's hard to find. The Samsung ICR18650-30A is also a 4.35V Li-Ion and is available from Orbtronic here (http://www.orbtronic.com/batteries-chargers/3000mah-samsung-icr-18650-30a-high-capacity-li-ion-rechargeable-18650-battery-cell), but the curve is notably taller for the LG cell based on the one curve I was able to find for the Samsung cell. That LG cell is really outstanding.
-
Nice find on the battery data page. Super battery. Looks like you're on your way to build another master piece of a mod :rockin smiley:
-
I got a message back from the eBay guy confirming they are the LG 18650 D1 cells. I ordered a few. I'll be doing some of my own tests with them.
-
Cool info. :thankyou:
-
So I thought I would draw up a quick enclosure for a couple of these cells to see how it would look in terms of size. Looks like it might be an okay. The dimensions would be 80mm x 40mm x 30mm. I still don't have the cells but they should be here soon. I'm actually excited about doing this one. The 1400mAh ones I did last time are great in terms of size, 80mm x 38mm x 19mm, but they just don't run very long on a charge, especially since I'm usually vaping around 20W.
-
So I thought I would draw up a quick enclosure for a couple of these cells to see how it would look in terms of size. Looks like it might be an okay. The dimensions would be 80mm x 40mm x 30mm. I still don't have the cells but they should be here soon. I'm actually excited about doing this one. The 1400mAh ones I did last time are great in terms of size, 80mm x 38mm x 19mm, but they just don't run very long on a charge, especially since I'm usually vaping around 20W.
That's a mighty fine looking case
-
That case looks fantastic. I love seeing CAD drawings like that.
-
It's kind of a fatboy, but I think it will be okay. Pretty amazing that I can quadruple the run time by making the mod 10mm fatter. Slight increase in width. The energy density is almost double for the 4.35V round cells compared to the LiPo flat cells. Also, the impedance is lower than a regular ICR cell. Between that and the higher operating voltage, two in parallel can output 40 Watts at minimum voltage.
Total impedance is 40m Ohms with two in parallel which would be over my limit of 30m Ohms, but the higher voltage increases my impedance limit to get them in just under the wire. Everything really falls into place nicely with these cells. Just have to modify the code as required and use a different charger controller. No other hardware changes. Kind of chomping at the bit to get them, but they're coming from overseas so it's going to take a couple three weeks. And, of course, they have to live up to the data I linked to in the OP.
-
Sounds like it will be a fun project! I was wondering though, why did you decide to have the counterbores of the screws on the face of the device with the buttons and LCD?
-
Yes, I have the same feeling about that. I just can't figure out a way to mount them on the bottom without the requirement to use really long screws. A deep countersink on the bottom would be a gunk trap. The 1/2" #2 screws I'm using were hard enough to find as it is. Also, putting the bearing load in the top would reduce the depth of the thread bite quite a bit. The top piece is only 6mm thick. With a black finish and the black oxide screws, it's not too horrible looking;
-
Just out of curiosity Craig, did you use the sheet metal feature for making the enclosures for your Powerbloks? I'm playing around with it right now and it seems way easier than messing with the Boss/Base feature.
-
I do an extruded boss/base, then a shell. Just extrusions, cuts, fillets, and chamfers after that. Haven't used any of the sheetmetal features. A base and a shell seems to be simple enough.
-
By the way, which charger controllers support this mode?
I know of MCP7383x line of chips. Anything else?
-
I like the TI charger controllers a lot. They have input voltage protection which is really good to have. Here's one for the 4.35V cells, up to 1A rate;
http://www.ti.com/product/bq24095
-
Damn, it's almost unavailable... too new for most of the suppliers ;) Anyway, I'm far from this stage of my development, probably when my circuit is ready it will be available already :)
-
What's wrong with the Microchip part? I ordered some of the 4.35V ones from Microchip Direct (MCP73833-NVI/MF). They were 85 cents USD a piece and shipped right away. Microchip has warehouses worldwide so it shouldn't be too expensive for shipping.
-
I mean this new chip from TI - it's unavailable almost everywhere... Microchip's ones are there :) (not to mention I like Microchip because of their MCUs ;) )
-
Microchip stuff is pretty good for the most part, but sometimes they do stuff kind of half ass. I like their ucontrollers, but you take some lumps with their horrible revision system using them. TI is what Microchip will never be in terms of getting things right the first time. Microchip's strong suite is good documentation and end-user design that is easy to implement. TI really falls down in the "easy to implement" arena.
-
Orbtronic stocks the LG D1 18650 now;
http://www.orbtronic.com/18650-li-ion-batteries-chargers/lg-d1-18650-3000mah-high-voltage-li-ion-battery-cell
-
This is what I love about building mods, getting stuff like this in the mail. It's way better than Christmas.
It's kind of a fat sucker, but with 6Ah, what can you expect.
(http://i1232.photobucket.com/albums/ff380/chblock/PB6000_case_zpscd83b1b0.jpg)
(http://i1232.photobucket.com/albums/ff380/chblock/PB6000_case_hand_zps301c60b8.jpg)
The battery pack fits perfect, damn I'm good :)
(http://i1232.photobucket.com/albums/ff380/chblock/PB6000_case_battery_zps3f94889e.jpg)
-
Merry Christmas.
Beautiful. Love the looks. :rockin smiley:
Fits perfect? of course it does. Design by Craig :thumbsup: I would expect nothing less
-
I would have said "OOOH shiny" lol but its not shiny. I think it could still apply here. Looks great Craig!
-
Yeah, hehe, it still won't be shiney after I finish it with flat black lacquer. At least the connector is shiney :)
-
Do you have to tap the screw holes yourself or can they come tapped from Solid Concepts? Just curious since I might order from them very soon.
-
They self tap. The nylon composite is really strong. I've seriously removed and replaced the screws a hundred times on the first one I did as I was fitting the parts. They hardly even loosened up. However, try to use the same thread track every time. I have little trick for that. Once I've done the initial thread cut by simply installing the screw, I back the screw out when replacing until it clicks down, then proceed to tighten it. That way, I find the original track. Use a 1.6mm diameter hole for the #2 screw threads, 2.2 diameter for a close fit on the floating section.
-
Where do you find the #2 self tapping screws? I'm having a field day trying to find them myself.
-
Yeah, they're tough to find. I got them here (https://www.fastenal.com/web/search/product/fasteners/screws/sheet-metal-screws/_/Navigation?searchterm=&sortby=webrank&sortdir=descending&searchmode=&refine=~|categoryl1:"600000 Fasteners"|~ ~|categoryl2:"600051 Screws"|~ ~|categoryl3:"600066 Sheet Metal Screws"|~ ~|attrdiameter:107900|~ ~|sattr04:^Flat$|~). They're not self-tapping per se, just regular sheet metal screws will do the job.
-
Internals are assembled, coded, and tested. Getting 5600mAh on the bench at 1C which is really good with these cells for a couple reasons. For one, capacity ratings are normally done at 1/5C and fall off with higher loads. For example, the Panasonic 3400mAh NCR18650B drops off to 3100mAh at 1C. Secondly, these cells operate at a higher voltage so when powering a DC-DC converter, that 5600mAh is actually closer to 6000mAh with a 3.7V battery.
Power limit is 36 Watts. Voltage range is 4.0 to 7.4V. Current limit is 8 Amps at 4V ranging down to 5 Amps at 7.4 Volts.
On to the enclosure now, that will take me a little while, not a quick task.
Here's the pics;
(http://webpages.charter.net/chblock/PB6000_internal1.jpg) (http://webpages.charter.net/chblock/PB6000_internal2.jpg)
-
Nice ;hubba;
-
Looks great as always Craig. Can't wait to see what it looks like in the enclosure. :thumbsup:
-
It's a beauty. Well done again. Lot's of power and battery life.
-
The specs sound very impressive. Nice work Mr. Craig
-
Can't wait to see this baby in action :popcorn:
-
Coming together, maybe finish it tonight. Here it is in the enclosure with the top off;
(http://webpages.charter.net/chblock/PB6000_top_off.jpg)
-
Finished this build today. It's a pretty hefty sucker, but it's tolerable and the run time is amazing. Here's some specs;
Dimensions: 81mm x 40mm x 30mm
Weight: 155g (5.5oz)
Type: Variable voltage
Voltage range: 4.0-7.4 Volts
Maximum power output: 36 Watts
Maximum current output: 5A at 7.4V to 8A at 4.0V
Battery capacity: 6000mAh
Battery type: Dual non-removable 3000mAh 18650 cells, LG brand (LGABD11865)
Charging: USB, mini-B
Max charge time: 6 hours 30 Minutes (USB charging adapter required)
Features: ammeter, wattmeter, ohmmeter, battery capacity gauge, temperature gauge, hit counter
Material: Nylon-fiberglass composite enclosure, 304 stainless atomizer connector
Here's the pics;
(http://webpages.charter.net/chblock/PB6000.jpg)
(http://webpages.charter.net/chblock/PB6000_hand.jpg)
Here's a shot of the LCD. That's from a run on my load bank at max output.
The box will run up a 1.5 Ohm atomizer up to 7.4V no problem, pretty stinking hot vape though.
(http://webpages.charter.net/chblock/PB6000_max.jpg)
This is the rear view in comparison to one of my 1400 boxes. You can see it's about 50% fatter.
Pretty amazing when you consider it has quadruple the run time;
(http://webpages.charter.net/chblock/PB6000_back.jpg)
-
Thats another awesome build Craig. Almost the same size as my DNA20 Brick, only about 12mm shorter in length.
-
Absolutely spectacular Craig. I see you added a temp sensor and a hit counter too.
-
Simply Amazing :thumbsup:
How does the temp gauge and hit counter work?
-
How does the temp gauge and hit counter work?
When the bottom button is pressed and held to turn the device off, an alternate screen momentarily displays with the last recorded temperature for the booster and the hit counter.
Temperature monitoring for the booster is required as a matter of safety. I figured I'd just a throw a table in code to translate readings from a thermistor located under the booster to a temperature display. I dont know how useful it is for an average user, but I actually find myself using the temperature gauge sometimes to see how well heat dissipates as a matter of design.
I use the hit counter quite a bit to gauge the longevity of my batteries and to check run times. I can tell how much they wear by how many hits between charges. It's not exact, but it does provide a good estimation.
There's a master reset button under the charge lights that can be pushed with a pen or other sharp object. It resets the hit counter and resets the voltage selection to 4.0 Volts. It also performs a master reset on the micro-controller. I actually use that often to jump to 4.0V for a dry burn so I don't have to scroll down using the "down" button.
Almost the same size as my DNA20 Brick, only about 12mm shorter in length.
This 6000mAh box does look like a brick. This one is a lot more like a brick than my 1400. I've been using it all afternoon now and I'm finding it is a bit heavy and cumbersome, but boy is that run time nice. With the little ones, I have to have them plugged-in a lot. Not with this one. I'm a heavy vaper and I use a good amount of power, usually around 18 Watts. So, I drain batteries pretty fast.
-
Great Craig where do I find those parts? I like the features even if it's pretty large would make a good home/car mod! :applaude:
-
Great build, Craig :thumbsup:
Where do you do your enclosures?
-
Great Craig where do I find those parts? I like the features even if it's pretty large would make a good home/car mod! :applaude:
fsors, I don't think it's large at all. It's the same L x W as the smaller version, just thicker. Still looks smaller than most mods.
-
yes it prob fit in a flatrate box if he wants to share Doh:
-
Thanks for the positive comments guys :thumbsup:
Where do you do your enclosures?
I use a 3D printing service with the SLS method. The company is https://www.solidconcepts.com. Now, this is expensive, not really cost effective to do just one, but I usually get a number of them at once then build a few boxes for others. Makes the cost not so bad, but even then it's expensive. My 1400 boxes worked out to $60 an enclosure, the 6000s were $80 an enclosure because I did them in a smaller quantity. If the cost is not an issue, there's no better custom enclosure unless you use CNC with metal and it's not any cheaper for small quantities, probably more expensive.
To use any of these production methods, you use 3D modeling software. I use a professional program called SolidWorks. I have an engineering friend who turned me on to a student version. You can also use programs like SketchUp which have a free version, but SketchUp can be limited since it's not really designed to make parts and assemblies.
If you want to really get into it, you can build your own 3D printer. You have a choice of the CNC subtractive type or the additive type that use an ABS filament. I actually have one of the filament printers and they work decent (http://www.solidoodle.com/). I could make an enclosure with it, but it's ABS plastic which does not have near the strength or heat tolerance of the Nylon GF used by SLS. However, lots of people use the inexpensive electronics enclosures you can get off the shelf. They're usually made of ABS.
Great Craig where do I find those parts?
My parts list fills a whole page. I use OshPark for my PCBs. They're actually pretty cheap, about $10 a piece which is amazing for a 4 layer board. Then there's the component soldering required to assemble the PCBS and the coding for the ucontroller. So that's a pretty big question. Is there something in particular you want to know about?
-
Thanks, Craig :)
However, lots of people use the inexpensive electronics enclosures you can get off the shelf.
And that's what I'm doing: using standard ABS enclosures (or aluminum ones).
-
Most people probably do. The aluminum ones are tough, but heavier. Believe it or not, that additional 2.5 ounces with the bigger mod makes more of a difference in comfort than the additional thickness. Plastic is nice because you don't have to worry about electrical insulation from the chassis. OTOH, aluminum saves you some wiring since you can use the chassis for ground. Aluminum is a great conductor.
-
I've been using this mod for a few days now and I've obtained some good measurements on hit counts. With the 1400 boxes I usually get around 320 hits on a charge. That may not sound like much, but keep in mind I'm usually vaping around 18 Watts. The higher power levels drain batteries pretty fast. 18 Watts drains a battery twice as fast as 9 Watts.
With this box, I'm getting around 1350 hits on a charge. So, it's getting over quadruple the run time of the smaller box. In terms of amp-hours, it works out to right around 6000mAh which is brilliant. These cells are higher voltage so they have more energy density. With higher voltage, that charge capacity gets better mileage. I'm probably getting right around the same run time as if I was using two of the Panasonic 3400mAh cells. On paper I expected that, but you don't know for sure until you actually try it.
I think I can safely say this mod has one of the highest run times of any e-cig mod out there. The only mod I can think of that would compare is a buck mod running two of the 3400mAh Panasonic cells.
Really happy with these cells from LG Corp. Fantastic performance. It's not just the higher voltage and the big capacity, they're also more efficient with only 66 mOhms internal impedance. The Panasonic 3400mAh cells have 100 mOhms. Though, nothing beats a LiPo. The flat cells in my 1400 boxes have 25 mOhms internal impedance. Still, 66 mOhms is pretty low compared to any other round cell.
-
Excellent vape time. More and more vapers are longing for longer vape time out of their batteries. Excellent batteries paired w/ your high efficient boost circuitry. Congrats :beer-toast:
-
Efficiency does become more noticable the bigger the battery. With this one, you can see each percent of the run time is 13.5 hits. The difference between something that gets say 90% at 20 Watts versus the 94% I'm getting at 20 Watts would be 54 hits. That's something you would notice.
-
Efficiency does become more noticable the bigger the battery. With this one, you can see each percent of the run time is 13.5 hits. The difference between something that gets say 90% at 20 Watts versus the 94% I'm getting at 20 Watts would be 54 hits. That's something you would notice.
Unbelievable difference in longevity and efficiency. Just wonderful
-
Hey Craig,
I was reading about batteries last night on one of the flashlight forums and came across the LG D1 18650 charts. I noticed that these cells are not typically meant for high-drain applications. They seem to be rated for a 2c max discharge. Yet you seem to be having no problem using them. How is that?
I personally would feel more comfortable seeing those new orbtronic pd2900 mah cells in there. I have 2 of them and they have been working wonderfully for me.
-
I'm running two in parallel. That gives me a max discharge of 12 Amps which is nominally 45 Watts. Normally it's not advisable to use parallel cells because large equalization currents can occur, but the cells are non-removable and permanently wired which eliminates the issue.
Otherwise, the taller discharge curve really simplifies things in terms of design. Mainly the fact they are fully discharged at 3.5V open circuit and they have low internal impedance for a round cell, about 60m Ohms. They behave more like a LiPo cell when comparing discharge curves.
I've been using this mod I built with them for a few months now. The run time seems to be holding up well so the cells seem to be fairly durable. That's always the question that takes some time to answer. Though, the LiPo cells definitely hold up better, but since the ICR cells have so much more run time and they don't need to be charged nearly as often, I still come out way ahead in terms of cell longevity.
-
Oh, I didn't comment on your second question. The NCR18650PD would be a viable option, it's probably one of the best cells to use, but the one ding is it has a rather steep discharge curve. To get the full run time out of it, I'd have to go with a 2.7V MCU supply which means a complete redesign of my circuit. My design uses a 3V MCU supply since it was originally done for a LiPo cell. They have the tallest discharge curve of any Li-Ion. The taller curve of the LG ICR cells allows me to use the same design.
The option was to use a cell with a taller discharge curve requiring only a change in the charger controller part number (which is trivial) or redesign the whole thing from the ground up. I used a cell that works for my design rather than a rework of my design for a particular cell.
Another advantage of the LG cell is higher Watt-hours. Since the curve is taller with the LG cells, I get more run time since it's directly proportional to a battery's Watt-hour rating. That's in part a function of cell voltage. Higher cell voltage means more Watt-hours which is where the rubber hits the road in terms of run time.
-
Hi guys
first i need to explain how and why i landed here:
i was checking http://lygte-info.dk/ for some specs for sanyo/panasonic discharge curves and somehow the best one was always the LG D1
LG D1? what the.... LG makes bats? Do they have have an own fab? from the data it sure looks like it!
ok, ask uncle G... tataa.. here i am! :-p
since i'm kind of drunk i couldn't compute what this forum is realy 'bout, but man that mod looks fine!
you're only talking about the cell and some screws, but what' s that board!?
i'm into modding flashlights, but as far as ecigs go i only do some caps cutting and rebuilding vivi heads.
i'll do some more reading tomorow, but it sure looks like there are some tech savvy people here!
;cheers; haha... even the smileys are raw here! :rockin smiley:
-
Hi GreenAir, welcome :)
I think this is probably the best place to find e-cig modders building stuff from the ground up.
Flashlights and e-cigs have a lot in common as far as power requirements. Higher end flashlights typically use 10 to 20 Watts over a range of voltages above and below the voltage of a single Li-Ion cell. So do e-cigs.
Yes, LG Chem makes Li-Ion cells. They've been making Li-Ion cells just as long as companies like Panasonic and Samsung. You don't see them much at the retail level, but the LG cells are used quite a bit in laptop computer packs.
The LG ICR18650 D1 is a really amazing cell. It has the lowest internal resistance of any ICR 18650, I've bench tested about 66 mOhms. Other ICR 18650 cells are typically 80 to 100 mOhms. That and the higher operating voltage means more power delivered to your appliance.
That higher operating voltage is both its advantage and its disadvantage. The advantage is the cell has higher energy density. In terms of energy, the ICR18650 D1 carries just about as much as a Panasonic NCR18650B even though the LG cell has 3000mAh compared to 3400mAh for the Panasonic cell.
The down side is you need a non-standard charger for the LG cell. It uses a terminal charging voltage of 4.35V compared to 4.20V for a standard Li-Ion. Removable round cell chargers like that are not available. Though, you can find charger controllers with a 4.35V terminal charging voltage then make a USB charger adapter. It's what I've done for the e-cig mod I use them in. The LG cells perform impressively.
Oh, that board is of my own design. It carries a 36 Watt booster, an MCU that controls everything including a display, and a USB charger. I've been building e-cig mods for a few years now and that board has evolved over several iterations. The one pictured is probably going to be my final design. I don't think I can improve the function or reduce the size any more than I have.