After 2 weeks out of action due to my last crash and hand incident, - http://forum.freeflysystems.com/index.php?threads/our-cinestar-8-bites-the-hand-that-feeds-it.804/ - I finally got our CS8 repaired (the bits that I could see), ran some tests on MKTools (motor tests etc.) re-checked all parameters and changed some too. Removed the motor start up safety switch and reactivated the stick bottom left and right for on/off. All seemed to be running as per usual. We had a mass-swim event taking place that we've been wanting to shoot and it was yesterday. We set up the CS8 as per usual, did 2 flights using our 4S 6200mah and all went well. I then swapped the low battery for a 4S 8000 mah battery, took off and about 20 seconds into the flight motor 7 went dead! Luckily I was high enough to not crash even though I lost some altitude when it happened. Manage to bring the CS8 back without too much trouble, landed and switched off. Motor 7 was non-operational. Only when I got back to the office and having taken a closer look I noticed BL7 on the MK PDB was fried. What could cause this to happen? Could this have been a latent failure from my last "hand" incident having perhaps taken a bump on motor 7? I've checked the flight log and all temps seem ok ... maybe I'm missing something? Would a faulty motor cause the BL to burn up? Could the BL burn up and the motor still be 100% - how can I test to see if the motor is dead? Any suggestions or help from vastly experienced crew on this forum would be greatly appreciated. Photos on BL7 attached, video of the flight and GPX file. (I've uploaded the log from the flight before the motor failure as well as the log from the motor failure flight, perhaps there were some warning signs in the previous flight that I missed) Much appreciated!
Sorry to hear that talk about bad luck. I know how frustrating these things can be I had a bl go out because the atmel Chip zapped. Chances are your motor is still good. I don't think the motor caused this I think it is a matter of electrical conductivity on the bl wires. In your case, not to sound abrupt, but your soldering job is not ideal, from the picture. You have all of your wires dangerously close overlapping each other. Even if they are not touching they can still cause an arc bridge and short out if they are close enough. I suspect this is what happened to your board. I am no solder expert by no stretch of the imagination. But I know a little bit to get the job done. Looking at your solder joints in general they are too "blobby" looking and are dull. This is an indication that your solder joints are cold meaning that they are not making a good electrical connection in general. I would have your center wire soldered so the wire pigtail Is centered going away from the board and your left wire exactly opposite of you right wire. In addition you will want to keep the heat on the solder and wire long enough for it to flow for a few seconds, while holding the wire in place with pliers or have a glove on. Once the solder has started to flow a bit then release the iron holding the wire steady in place for a few seconds. The joint should have a nice silver luster to it versus a dull sheen. As for desoldering that bl good luck it's a bitch. I would suggest taking it to a shop with someone who has the skills to do it. You will need a good hot desolding gun with a desoldering plunger. If not careful you can easily delaminate the board. Ask me I know :-/ On a good note at least you saved it, reaffirming my belief in the 8 rotor over the 6. Good luck
Hey... I just had almost this EXACT issue... http://forum.freeflysystems.com/index.php?threads/help-what-the-heck-happened-here.848/ The cause? Who knows... I'm still stumped. Ended up buying a whole new PDB. Yea! $800! Can't wait until someone comes out with a more robust FC and PDB solution... Perhaps not quite as exposed to the elements... Best, Ben
Ben I believe your issue to be the same. You had some exposed wires that looked real close to the adjacent wire. 20 to 40 amps of current is a lot of electron flow. That is enough to cause an arc if two wires are near each other. That electrical arc is essentially the same thing as if the two wires are physically touching. It can and will cause a short or at least a disruption in the free moving electrons to cause electrical components to burn out. I agree with you I would like to goto a better type of system. There are too many points of failure with the MK PDB. Eventually I am modifying my FC to plug into separate escs like castle ice, using a 12c to PPPM converter negating the need for he MK Board all together. I zapped my bl micro controller that little atmel at168 chip that you see to the left, with a static discharge when I was soldering the led wires. I have an ESC work bench with grounding strap. That day I was complacent and I made an 800 dollar mistake by not simply strapping on a grounding wrist strap.
Hi Shaun, Thanks for the feedback ... I agree the soldering is a bit sketchy! Since soldering the initial build of my CS I have had many more hours of practice with other bits and pieces! My initial thought process when I saw the burnt out BL was not to replace it but to purchase a complete new PDB and BL's and start the soldering from scratch ... a huge price difference compared to just replacing the BL but it could alleviate any glitches due to beginner soldering in the future and the potential loss of the entire CS. Also much easier than trying to remove the current solder. I also presume that the BL on my PDB is avs 1.0 and not the newer vs2.0 ? So in replacing the fried BL I would have to purchase the older version? Excuse my ignorance but my lack of knowledge is directly related to my experience with the MK setup. Thanks for the soldering tips ... will definitely put them to practice! I was quite surprised how well the CS handled with one motor out, after the initial dip in altitude it was no problem bringing it back and landing ... definitely helps having the extra 7 motors!
Could it be worthwhile to use some "liquid electrical tape" on the motor wires where they are soldered to the PDB?
Shaun ... that was going to be my next question ... seeing as though I was thinking of replacing the entire PDB and BL instead of just the damaged BL, would you suggest that perhaps it a better idea to convert to seperate ESCs etc. now? It was something I contemplated doing on a new build or converting our current CS, but it was a "way in the future" plan ...
I believe you have the current board. I think that the first iteration, the BL's were Orange as well as most of the first generation MK stuff. I would say if you are looking at going to an entire new board versus trying to salvage your current one. I would suggest going with a completely different ESC setup. You will need to get a I2C to PPM converter and 8 3 prong servo wires. So far it seems that Caslte ICE 50 Amp ISC's are proving to be reliable. There is minimal soldering involved. I believe the only soldering you will have to do is from the I2C port on the bottom of the FC board to the PPM converter. http://mikrokopter.altigator.com/i²c-to-pwm-converter-for-standard-controllers-i2c2pwm-p-40611.html No more pico blade Molex connector. You will have to build a power distribution block to feed the battery power to each ESC I believe people are hot gluing 8 Deans plugs together for this. there will be soldering involved similar to the MK PDB. You will want to make good solder joints and heatshrink evrything! You do not want a arc bridge, it can stii happen in this setup. There is a whole discussion on the heavy lift build section of this forum. You will need to tap power from the battery to your FC. Someone mentioned using a 12V BEC in the middle, not sur if this will be necessary for the 4S. Of course I have not done this yet so I maybe full of it I would hit up the people who have, if you decide to go this route. In hindsight if I had known this was a possibility when I trashed my first board I would have jumped on it. You will have a system more ready for future upgrade. With your system you will have to do some further mods when converting over to a bird that has bigger motors props and can handle 24 volt batteries.
Probably not a bad idea, or at the least heat shrink the wires as close to the solder joints. In either case probably still want to keep the wires away from overlapping on other connections as much as possible. Not sure how much isolating properties that Liquid stuff has. If two wires are close together regardless could still be enough to allow for current to jump out, not saying it will though. But better to asume it can versus it cant. The question is how good is that stuff when the BL's get to high temperatures?
Probably not a bad idea, or at the least heat shrink the wires as close to the solder joints. In either case probably still want to keep the wires away from overlapping on other connections as much as possible. Not sure how much isolating properties that Liquid stuff has. If two wires are close together regardless could still be enough to allow for current to jump out, not saying it will though. But better to asume it can versus it cant. The question is how good is that stuff when the BL's get to high temperatures?
Just for a bit of clarity, most of the CS setups run about 8-10 amps per BL, total in the mid 60 to 80amp range. If you pulled 20-40 amps the BL's would be toasted and probably other bits and pieces. Sebastian you have the BL 2.0's. If you are ultimately going to do heavy lift (Epic, C500, etc) then you will need motors, escs, new power distribution, etc. That said there are probably thousands of copters successfully flying the MK setup (CS, MK, Droidworx, etc). Not to say that there aren't failures but they do seem to be rare and are usually due to the items being discussed on the forum, like wire whiskers, pins that break due to vibration, etc. Going the individual esc route has just as many failure points I would guess as the MK stack. Added to that is a much lower level of support since you have such a custom solution. All that being said Josh, Brad, myself and others are having some success with the I2C setup but it is not without its own set of growing pains.
Ill agree 20 amps probably high but not out of the relm of reason. Mikrokopter.de claims 35 Continuous peak 40A so i am no too crasy. Probably will get a BL limitation most likely from reaching those levels. I think my point is still valid in the sense that if you have two wires near each other as in the picture with the right potential and current flow, can cause an arc, and disrupt the smooth flow of current. Looks like it was pretty spontaneous. I looked at the both GPX files the second one shows the temp on 7 higher than rest for the same current draw. But the first file invalidates that hypothesis in the sense that 1 has a higher temp for similar current draw, making the data fairly inconclusive for prediction. There can be many theories I suppose to why the inconclusive data. It could be that this was a issue that was slowly causing deterioration to that BL until it reached a limit were it was not going to take anymore stress from disruptive current flow and finally broke down. As to the Seperate ESC discussion having not done one myself I yield to gary and those that have
And the component that poofed where you have the crater is a capacitor. I've had them pop on other components. Usually a power overload or reversed polarity. And they don't leave much behind when they go. Shaun you are right on the specs of the BL's. I think what folks have found is that they run out of temp before they run out of amps. Summer for me was avg of about 70 amps but the temps were at or slightly above 100. A reason to change to more robust individual ESC's just for the temps.
I have a couple those CAP's do that back in the days of setting electronics on fire as an early engineering student. We were taught all the theory on making circuits on paper with ideal Voltage and Current sources. When it cam came time to implement it in a lab they failed to mention things like Electrolytic capacitors having polarity. I created a Wheatbridge Diode VR with a couple Electrolytic caps as a filter. Connected the scope and the AC power supply I got a nice flat line for about two seconds and then the scope started spewing out garbage. Looked over at the circuit board and saw one cap bulged out like basketball and smoking. The little Chinese lab TA tells me "You have capacitor back word." lesson learned on that. On a serious note i suppose that is good way to tell if something is going wrong, with the board. If the Caps start bulging like a soda can left in the freezer for too long, its probably a good indication that something is causing a short somewhere. I would be interested to see a pic of the bottom of his MK PDB to see if other BL's caps are having an issue.
There's definitely a silver lining to this story, and that's the redundancy of having 8 motors paid off big time. Replacing one BL, or even the entire PDB, is definitely preferable to the disaster that could have resulted if the copter either veered into the crowd of people below or into the lake to the right. It's a bummer that it's going to be grounded for a while, but kudos to you for having the presence of mind to get 'er back on the ground quickly and safely!
Hi Shaun, I'll take a pic his evening of the board from below. As far as can remember there were no burnt spots ...
Not expecting burnt spots. Just curious if any other capacitors are showing any signs of potential failure on other BL's. If you have any other caps that are going bad sometimes you can tell if they start to bulge out.
Shaun, Herewith a pic of the underside of BL7. Can I ask you ... or anyone else in the know ... to have a look through the 2x GPX files I attached at the start of this thread to see if you can see anything obvious that stands out that could be of concern! As always, all the effort is much appreciated! Thanks.
I'll try take a clear enough pic of the entire top and bottom of the PDB for scrutiny and post it shortly.