Hello everyone, I wanted to start a new thread here about anti-spark plugs, which is discussed here: http://forum.freeflysystems.com/index.php?threads/alta-death-spiral-and-crash.7322/page-6 Instead of ruining Adam's topic, it would be better to discuss it here. Here are some notes from that page: * There is a speculation that DJI A2 and WKM is vulnerable to antispark plugs; they can't seem to boot properly, as Radolf tested it. In addition to this, Globe-flight.de puts note on A2 controllers which states not to use antispark plugs with them. * DJI itself, uses AS150 plugs on their S1000 ve S900 ARF systems. *Kopterworx uses Herkules' antispark systems on their own branded RTF systems and they are working good, as Kopterworx notes this in their post. * I have been running AS150 plugs on MK systems with BL Ctrl 3.0 and Herkules III for quite long time. No issues so far. Tabb mentioned this antispark plugs in response to Adam's crashed Alta as : I would like to hear opinions on this antispark plugs.
Does anyone know the specifics of the technology used in these anti-spark connectors? Normally, the sparks that occur when you first connect up the LiPo are caused by the capacitors in the ESCs charging up -- and for that first few milliseconds those capacitors "look like" a dead short across the LiPo so you see a very large in-rush current, hence the spark. My understanding of how the anti-spark connectors work is that the initial contact is made with, not with the main connectors but by connectors that have a resistor in-circuit. The effect of this is to reduce the in-rush current, allowing the capacitors to charge up more slowly, and thus eliminating the spark. See e.g. http://www.espritmodel.com/jeti-afc-anti-spark-connectors-5-5mm-150a.aspx For the rest of the avionics drawing power from the LiPo, this reduced in-rush current would appear as though the DC voltage from the LiPo is being gradually increased (for electronics geeks, think charging an RC circuit). See http://www.electronics-tutorials.ws/rc/rc_1.html for the morbidly curious. What's not clear to me is why this more slowly rising voltage would create a problem for the other avionics -- it just would mean they too would experience a lower that normal in-rush current as the voltage rises. Is it that these other electronics go into some kind of confused, non-deterministic state? Andy.
some ESC manufactorers telling the antispark can harm their electronics if the electric resistance of the antispark is to high. may this could also create a problem with the initialization process of the FC? but i really don't know. all i know is that the (Jeti) Antispark i used did that problem i had with my wkm. the electric resistance of the Jeti is 30 Ohm.
and i wonder why freefly didn't install an antispark when there is no problem with them? it is not expensive - just a resistor... that's the reason why i didn't install one on my own. it would be great to hear what the freefly techs think about antispark.
[ disclaimer: I'm not an engineer, although I do own the correct headgear ] If you were concerned about sparks on the battery connectors, couldn't you carry around a very small 24V battery (maybe 250mA) and plug it into your copter before connecting your big LiPos to charge up the copter's capacitors? I've noticed that there is no spark when I attach my bench power supply when it's set to a couple amps, and when I subsequently connect a big LiPo, there is also no spark. Maybe one of the 'E' engineers here can answer that. Seems it'd be a much safer way of avoiding sparks.
@SteveM: The aspect of the 24V battery that stops the spark is that its internal resistance is likely to be much higher than that of a LiPo, and thus is current limited -- which is also the reason why the bench supply doesn't cause a spark when limited to 2 Amps. So you are correct, you can prevent a spark by limiting the current -- which is what the anti-spark connectors do. I'm still struggling to understand why having a voltage that increases over a slightly longer time frame (like a few hundred milliseconds) would cause flight controller electronics and/or ESCs to have problems. The other technical question i still don't understand is, if the antispark connectors operate by having a resistor in circuit, doesn't that current limiting effect persist even while the aircraft is flying??? Would that not serve to limit the effective current that the ESC's can pull from the LiPo as the resistor would appear to be added to the internal resistance of the LiPo (it's in series with the LiPo)? Perhaps *that* is the problem that causes grief for the ESCs? Or do the resistors cause a voltage sag under heavy load? Either way, if someone can post an MK flight log or ALTA flight log where anti-spark connectors have been used, we can examine the flight log and see if there are voltage sags. The ALTA flight log would be more persuasive as a data row is written every 40 milliseconds -- whereas the MK flight logs are only written every 500 milliseconds. Andy.
in a german forum there is a thread (sadly in german) where a distributor of dji is telling about the problem and that dji knows about it. He had several clients with problems (all with antispark) and together with dji and log data they figured out that there is a relation between the antispark and the problems. One can have hundreds of flights without any problem. i can remember that the copter flew like it had a bad compass data, like one half wanted to fly in opposite direction than the other. sounds funny but it wasn't may there is a bad compass data caused by the antispark? i don't know... you won't get any log data from dji nor information but globe-flight is the biggest dji reseller in Germany and recommend not to use any antispark with wkm or A2. Better safe than sorry... i wouldn't use an antispark until this is effectively solved. and i wouldn't use an antispark with ALTA too untill freefly can tell if there is a problem with it or not.
Rolf: Could you post the link to the German thread. Google translate will allow interested people to get some idea of the messages. Thanks Andy.
http://multicopter.com.de/index.php?topic=224.0 but i would trust an antispark more than the google translation
Thanks! If you paste that URL into translate.google.com and set to translate from German to your preferred language, and click on Translate, then it will translate the entire web page. The translation is a little bit "odd" as Rolf says. From the same literary school as "On the balcony have been I since 10 am who soon do I become a sausage?" But you can get a general understanding. Andy