Dixie vspor Kennesaw ga. Dave got a few today probably sold out in the next 40mind
- Status
Ol' Blue gets a 40. I just built this with a dna30 in August. Not that old unless we are dealing with vape devices it seems.
I started out trying to build a series coil in the attached Veritas dripper. I didn't plan that well at all and ended up with a single coil. Whatever.
The single coil works good in it anyhow. Ended up with 30awg Ni200 at .18Ω./470°/23W. Lost count on the wraps (9-10?)and just got it done. The coil dropped to .16Ω after a little use and stayed there. Works perfecto!
We are starting to realize the real value with the chip. Non-critical coil building skills and no more lungs full of burnt puke. I've run the cotton down to a dry surface with a very slight "too dry taste". It's still as clean as ever. My tongue is no longer getting pelted on occasion with damaging hot drops of liquid. Basically worry free once the temp and wattage is dialed in. Two things: Set it to the heat you want without being too aggressive. Set the wattage to an area where it's not constantly going into protection. I can see cotton wick lasting weeks.
I started out trying to build a series coil in the attached Veritas dripper. I didn't plan that well at all and ended up with a single coil. Whatever.
The single coil works good in it anyhow. Ended up with 30awg Ni200 at .18Ω./470°/23W. Lost count on the wraps (9-10?)and just got it done. The coil dropped to .16Ω after a little use and stayed there. Works perfecto!We are starting to realize the real value with the chip. Non-critical coil building skills and no more lungs full of burnt puke. I've run the cotton down to a dry surface with a very slight "too dry taste". It's still as clean as ever. My tongue is no longer getting pelted on occasion with damaging hot drops of liquid. Basically worry free once the temp and wattage is dialed in. Two things: Set it to the heat you want without being too aggressive. Set the wattage to an area where it's not constantly going into protection. I can see cotton wick lasting weeks.
Attachments
Last edited:
I had the same issue. None of my atty's would fire because the cheap non-adjustable 510 connector was too deep. Sent it back asking for a refund.
Last edited:
Imo for a 250 dollar mod...that should not even be an issue.
(sent from Mars, it's not as red as I imagined.)
I totally agree! They said they would replace it but my poor brother has to wait for it to come in all over again. I'm just baffled they don't just put in an adjustable pin, I'm sure it doesn't cost that much to have a putting loaded pin added to the design.
Me too....seems it would be an inexpensive design fix....your poor bro....tell them you want a refund and order an xpv? I seriously love mine...and it is fifty dollars less. Five days lead time and that little wheel makes settings a breeze....spring loaded pin ...! My ruzgar has a short pin and it vapes like a champ on it.
(sent from Mars, it's not as red as I imagined.)
Last edited:
The reason you need to limit the power setting is that, unless Evolv have implemented a very good control algorithm that can learn to tune itself, the temperature will overshoot. Take a look at this graph:
Hmm interesting!
But does this problem actually apply to this problem? If you control the "speed" of the temperature change directly with the watts instead of an "acceleration", you can just set the power to 0 watts and the curve should flatten immediately. If you where to measure the temperature externally you would have a lag until the heat produced reaches the temperature sensor through some kind of medium, then you could get these overshoot problems, but since you directly measure the resistance? Not sure really. Of course you do have a constant loss of heat that you want to compensate, so maybe that introduces this type of hysteresis error again into the system.I have never implemented this but I guess if you look in a textbook you'll find some kind of self balancing algorithm for these types of problems? I bet these are the type of problems that the CTO of evolv loves to nail perfectly
Man, this just sucks. I figured they would have addressed this by now. This is the only thing it needs to go from problematic to very good. You're not even allowed to pull the lid off of the chip to push it back up while under warranty.
Finding someone locally with all the right tools and skills to install a fatdaddy connector is probably a brutal task. I can vouch for the durability of this configuration however. I've wrenched the hell out of it with no problems at all. No internal leaking.
Attachments
Hmm interesting!
I have never implemented this but I guess if you look in a textbook you'll find some kind of self balancing algorithm for these types of problems? I bet these are the type of problems that the CTO of evolv loves to nail perfectly
Dejay.
That graph is a standard PID Control example and the Green trace is the optimum response curve right out of a "Closed Loop Controller Handbook". It doesn't get any better than the "Green" trace. The "Green" tuning gets to temp the fastest with the minimum overshoot and secondary undershoot.
I am amazed that Evolv has not put out some recommended settings for their board. They have to know exactly what the best configuration is.
I made a post stating that I would love to know what their approach to temperature control is which I am sure they will never tell but I am sure they could give some typical setting without giving away the proprietary design.
I haven't been able to even order one yet "Out of Stock" is the code word. But is we are going to post tuning results i.e. what works and what doesn't we are going to have to state wire ga. and length. The mass of the Ni200 has a lot to do with the operation of the controller. I assume that mass calculation is part of what is done when it thinks the atty has changed or when it asks you if it has changed so it must be important to the control algorithm. Once calculated they then use data from a look up table in conjunction with monitoring the change in current flow (delta resistance).
There is one other factor that I have no idea how they deal with and that is the internal resistance of different attys i.e. I have three ZAPs (Z-atty Pro) and they use a spring loaded positive connector. One reads (on my Fluke Multimeter) 0.1 the second is about 0.2 and one I need to take apart and clean that reads 0.35 to 0.4 ohms. That is a substantial portion of the entire Ni200 coil resistance. "Just one of those things that make ya go Hmmm..."
PS - I don't even remember what I started out to say. The DNA-40 software is pretty darn smart or someone is pulling the wool over our eyes.
Dejay.
That graph is a standard PID Control example and the Green trace is the optimum response curve right out of a "Closed Loop Controller Handbook". It doesn't get any better than the "Green" trace. The "Green" tuning gets to temp the fastest with the minimum overshoot and secondary undershoot.
I am amazed that Evolv has not put out some recommended settings for their board. They have to know exactly what the best configuration is.
I made a post stating that I would love to know what their approach to temperature control is which I am sure they will never tell but I am sure they could give some typical setting without giving away the proprietary design.
I haven't been able to even order one yet "Out of Stock" is the code word. But is we are going to post tuning results i.e. what works and what doesn't we are going to have to state wire ga. and length. The mass of the Ni200 has a lot to do with the operation of the controller. I assume that mass calculation is part of what is done when it thinks the atty has changed or when it asks you if it has changed so it must be important to the control algorithm. Once calculated they then use data from a look up table in conjunction with monitoring the change in current flow (delta resistance).
There is one other factor that I have no idea how they deal with and that is the internal resistance of different attys i.e. I have three ZAPs (Z-atty Pro) and they use a spring loaded positive connector. One reads (on my Fluke Multimeter) 0.1 the second is about 0.2 and one I need to take apart and clean that reads 0.35 to 0.4 ohms. That is a substantial portion of the entire Ni200 coil resistance. "Just one of those things that make ya go Hmmm..."
PS - I don't even remember what I started out to say. The DNA-40 software is pretty darn smart or someone is pulling the wool over our eyes.
Maybe internal resistance is the reason for the +/- 10 degrees thing? Don't catch me lying...lol
There is generally a heat reservoir involved. That would either push the temperature higher even though you'd cut power or, more likely in this case, drop temperature faster than you expect. So the first peak is at the right temperature but it settles far below it. You need the control process to choose exactly that power which maintains the temperature, and that power requirement changes depending on how hot the wick is and whether someone is sucking on the drip-tip.Hmm interesting!
I have implemented several of these. My customers tend to want to run an 8 litre engine with its unpressurised coolant at 98C. That's maybe a quarter of a megawatt input and two degrees margin of error. Get it wrong and they have to abort the 3000-hour test and get out the mop and buckets. I run that control loop at the same speed as Evolv -- 10 Hz, which makes me think they are running too slow to have very tight control. Brandon said they control to within 10 degrees and while most of that will be measurement inaccuracy, a good part will be control-loop lag. They may have an adaptive algorithm in there -- methods do exist to self-tune a control loop -- but if the control was tight the coefficients would vary depending on how hard the vaper was sucking. There's a limit to how tight they can make it.I have never implemented this but I guess if you look in a textbook you'll find some kind of self balancing algorithm for these types of problems? I bet these are the type of problems that the CTO of evolv loves to nail perfectly
A dry coil will start to glow in a couple of seconds. A glowing coil is 600-1000C. So the temperature increases at around 60C every tenth of a second. It would be near impossible to control that to within 10 degrees with a 10Hz control-loop. A wet coil will heat and cool slower, bringing into the realms of possibility, but a power level that was much too high for the coil would almost certainly overwhelm the controller and produce large over- and under-shoots.
Last edited:
Dixie had a few, I called Big Bang last night and they said it will be a few weeks for them! Thank goodness for Want to quit and him being able to grab me one while I stayed home with a sick little one!!! Thanks buddy!
I have my dna40 set at 500 degrees; its a single coil atomizer .15 ohms 30g Ni200 about 13 wraps 2mm ID contact coil.
From watching the LED the watts fluctuate but peak at about 17w.
I have it set at 22w.
I like this setting because it seems hot enough for good response, and never seems to over cook the juice (not sure though) since I get optimal flavor every vape, which is something I can't say about my other non TC devices when chain vaping.
Any comments about whether this is optimal (per the graph) or not?
From watching the LED the watts fluctuate but peak at about 17w.
I have it set at 22w.
I like this setting because it seems hot enough for good response, and never seems to over cook the juice (not sure though) since I get optimal flavor every vape, which is something I can't say about my other non TC devices when chain vaping.
Any comments about whether this is optimal (per the graph) or not?
Bass,
You may be correct about the +/- 10 degrees. But my guess is that holding within 10 degrees is a pretty good trick even if the atty was a perfect zero ohms.
If you think about it the DNA - 40 is like controlling your house temperature with a digital thermostat that has no temperature sensor. Now in my book that rates right up there with... well I can't think of anything but it is truly a marvel.
PS - If I can't get my hands on a DNA 40 Vapor Flask pretty soon you guy are going to drive me nutz (OK... more nutz) with envy.
Bass,
You may be correct about the +/- 10 degrees. But my guess is that holding within 10 degrees is a pretty good trick even if the atty was a perfect zero ohms.
If you think about it the DNA - 40 is like controlling your house temperature with a digital thermostat that has no temperature sensor. Now in my book that rates right up there with... well I can't think of anything but it is truly a marvel.
PS - If I can't get my hands on a DNA 40 Vapor Flask pretty soon you guy are going to drive me nutz (OK... more nutz) with envy.
It is a marvel and I love it!!!
I am setting temp to protect the wicking too. Been vaping at 400 degrees (burn point of cotton is 410 and I love cotton) set for 30 watts but I have never seen it climb above 6 to 7 watts before the coil is at 400. Man, the battery life that way is insane. I tried setting temp to 525 at 22 watts just to see it hold watts while the temp rises. For me personally the vape was too hot and sticky. I am certain there are people that like it that way, I am not one though.
Even at those settings the vapor production is great.
Thanks for the info, playing with big toys sounds really cool
I would imagine them running at far higher speeds? I guess they do use rather tiny and cheap microcontrollers but shouldn't they still be in the megaherz range? For example arduino controllers are quite fast for the 8bit stuff they do.
BTW that graph is purely for illustration of the general problem, right? It doesn't have any relevance for building on a DNA40, or does it?
You just tossed that in to make me feel even worse.
I am not into hot vapes either but I do like a good Plume.
- Status
