Sorry Josh, I should have been more clear..what's your stick position in hover? Andy's forensic work somewhat confirm my suspicion...a little slim on thrust.
Josh: I don't see any outlier data other than the BL Temp on Motor 3. I was expecting, based on your diagram, to see that #7 would also be working as hard (and as hot) as #3, but it's not. Is there anything obviously wrong with motor spin directions? I had it in my tiny brain that the lower motors contra-rotated with respect to the top motors, but your diagram suggests that's not true. Looking at the image you posted in message #1, the near right hand motors have the props set up for contra-rotating (if the red LED boom in the right foreground is #1, then motor 1 is set up for CW, and motor 5 is set for CCW). I've not really thought much about the two motors/one boom configuration so take what I write with a grain of salt (or sand!). Andy.
Yes, what Andy said. The lower motors should spin in the opposite direction in relation to the top motors. Same as the attached diagram, although wkm, it's the same for MK.
Oh sorry Howard. I am hovering right at mid-throttle. maybe 51-53% but nothing crazy. Andy, The rotations on the diagram are for the top motors only. The bottom motors counter rotate. Sorry I forgot to mention that. I just did a visual inspection on all the BLs and don't see a blown cap anywhere. I powered the copter back up and waited to see what the lights would do on the bottom and it's actually BL 1 that is having the issue. I am getting a Missing BL error in MKTools and the RED LED lights up on the bottom when all the others go green. I then re-inspected just BL1 and still can't find anything blown. Ill try and post pictures later this evening for your eyes as well. Also in the BL control error log it counted up to 128 on powerup for BL1. If you looked at the BL Found screen it was showing 2-8 but not number 1. I also wasn't getting any temp data for BL1. Josh
Unless it is one of the SMD caps that you see charred it might have popped a large cap but it is internal.
In my experience when one of those electrolytic caps go, they go out in style and very evidently -- but what Gary says could well happen. Certainly all the evidence is pointing at BL1 -- but that doesn't explain why the motors (and the BL-Ctrls) are running so hot -- it's almost as though the system as a whole is not happy -- some out of phase issue with the motor power? I'm speculating furiously -- but why would they all run hot. Does that point to the Mixer setup? What payload were you flying Josh? Anything? It should be able to lift its own weight without breaking a sweat. Just for grins, could you configure it as a quad, disconnect the lower motors and just see if the motors/BL run cooler -- it might not get off the ground but it would give you a clue as whether it is the mixer. Andy.
I was flying a 5dmkii on 3 axis. Ill try the quad thing and see what happens. How would I test the BL to see if it was a cap that went? Josh
Hi Josh: OK. So you've got a pretty sizeable load on there -- but still, your other Cinestar doesn't run with motors that hot, so why should this one? As for checking the cap while it's still in-circuit, that's a tough one because other elements of the circuit will interfere with your test. I'd start with a good visual inspection -- any sign of puffing up, discoloration, funny smell? If nothing shows visually, then I don't think (because of the other components that you can really give it a good test while it's attached to the PDB I'm afraid. Of course, if the red LED is on, you know you've got a problem -- the CPU on the BL-Control will turn that LED on for a good reason...but it's hard to know specifically why. I've checked the schematic at http://mikrocontroller.com/files/BL_Ctrl_V2_0_Doku.pdf but I cannot even find the two 25 volt 330 microFarad caps on the schematic -- it's quite late here, so I may just be looking and not seeing. They might be C21 and C9 shown in sector A 4/5 on the left hand side -- the designation is RAD330/16 for both of them. (Maybe RAD means "radial" or cylindrical?) If that is the case they the + side of both caps are connected to the same place so trying to test one will end up testing both. It looks like they're being used as part of the voltage smoothing circuit from the TA7805 5v regulator. The regulator along with the chain of other caps, provide Vcc -- the main drive voltage for the MOSFETS -- you can see Vcc (it looks like UCC on the schematic) on the top connector of the MOSFETs. Not sure that this all helps much, I'm afraid. Andy.
I was doing both. Since I had the new AH sensor installed I was trying to put it through it's paces as well. Josh
Andy, If I were to remove both caps from the board then I can test them correct? I've never tested a CAP before so Im not sure what I would be looking for. Ill ask Google and see what she comes up with. Josh
That would be the best way -- normally you can test a cap by setting a Volt-Ohmeter to measure Ohms. What you should see if you connect the Red lead of the meter to the +ve side of the cap (and the Black lead to the negative side of the cap, of course) is a low resistance, increasing rapidly as the cap charges up from the small voltage in the meter used to measure the resistance. A fixed low or high resistance means you're the proud owner of an ex-capacitor. Andy.
This test only works for electrolytic caps of some number of microFarads, for the picofarads range it doesn't work, as I recall. Also there's this: This is the slightly more complex way: http://www.ehow.com/how_6183832_test-capacitor-digital-multimeter.html Any EE's care to comment? Andy.
I have found what I hate the most... removing these stupid caps!!!! Even with a solder sucker it is a PITA!!!! Josh
I didn't want to psych you out by telling you that beforehand, Josh, but yes. Solder sucker and some solder wick might help. But either way, because the cap leads are part of the structural mounting of the BL-Ctrl, it's not easy. Andy.
BL1 is fried.. thought it was just the caps but just replaced those and it's still not showing up and the thing is SCORCHING hot even though it's not ever running a motor. Josh
Ok guys, Super confused now, I have replaced the BL1 with a known working used one. I have powered it all up and the LEDs are all green on the PDB however in MKTools it's still showing that BL1 is missing! If I look at the error counter it's at 127 as well. There also isn't any temp data from the BL. I plugged a motor in and during startup it spins the motor/makes the normal chirping sound but still shows ERROR 18 in MKTools. I also have no response from the motor when I do a motor test. I have re-flashed the FC with 90h firmware and after it does the reset it's still showing "MISSING BL1" in the flashing window. What would cause this? Is the actual PDB board fried too? Im confused. Josh
Hmmm. That is really puzzling. I don't know whether the BL-Ctrl does that start up test autonomously or under the control of the FC board....but what you're observing suggests it's autonomous and it seems like the BL-1 is not talking to the FC. Can you double check that you have the addressing on the BL-1 correct -- if for some reason it wasn't, it could produce the effect you're observing. If it's not the addressing, then, I presume that you cannot use MK Tool to run the motor test on motor #1, right? So working backwards, if the BL-1 LED is green, and you're sure the addressing is correct, then it has to be something in the data path between the BL-1 and the FC. There's not much in between other than the PDB, I'm afraid. The I2C bus comes down the Molex cables, and it's shared by all the BL-Ctrls -- so the fact that BL2 - 8 are not being flagged as missing suggests the I2C bus to the PDB is good. That means it has to be something on the PDB, or the I2C connections to BL-1. Andy.