SparkFun Forums

[Flydigital]

I have installed the Atto Pilot 180A version into a 6s multirotor setup. I want to use it to measure current and sent it to a telemetry port on the Rx.
There are 3 pins out GND, V and I
GND + V works fine and gives me approx 1.5 v reading corresponding to around 24v which sounds right. (The max reading is about 50+v which steps down to 3.3v)

GND + I however has a reading of zero. I think the atto pilot is supposed to convert the current reading into a 0-3.3v signal to correspond with 0-180A.

It's not working so I'm looking for any suggestions.

Note: My soldering looks pretty bad but I have done continuity tests which appear to be ok. No shorts and the multirotor powers up fine.
Can bad soldering or overheating while soldering compromise any parts?
Thanks for any help.

[Valen]

I think we need better pictures. The wire to Iout and it's ground is covering the shunt resistor, which could potentially be shorted by the other red wire on the right input side (can't see, need another angle) Then you can't get a current output.

The solder joint of the same red wire does seem like a flaky one, barely touching. But if the multi rotor power up then it must conduct some current.

The red wire on the output side does have a few curled strands coming dangerously close to it's ground wire. A bit of wire-bending and you've got yourself some sparks I'm afraid. (But this could be fun, right

I don't suspect you burned any parts. The INA169 current shunt amplifier is no where near your soldered spots. Though you way want to check for continuity between the Iout solder spot to the via on the other side of the current shunt, near the black marker dash. It goes to the resistor or capacitor that is nearest to the INA169. That trace or soldering hole may have de-laminated with too much heat. Again, hard to see because the wire is obstructing the view on the solderjoint.

[Flydigital]

Hi Valen,
Many thanks for your help. From what you say I think I can identify 2 issues with continuity. I have tried to take better pics but am lacking fine enough close ups.
If you see the picture of the front I have marked the points that I think you refer to.
Points A to B there IS continuity which I think is not right?
Points C to D there is NO continuity.

Note on the underside there is a bridge on the red wire which might not be right.

So if you can confirm I will separate A to B and build a link from C to D.
Hopefully that will help?!

[Valen]

The solder between A and B should definitely be removed. That resistor has 4 connection points, neither of which should conduct to each other.

You may not get continuity between D and C if you try to probe the hole at C. It is likely covered by a nonconductive coating. Instead put one probe on the solderpad of the resistor/capacitor on the other side. As it should be going to it.

[Flydigital]

Yes, at first I got no contact so I tried the reverse point of C and D and still no good.

[skimask]

One of the first things that needs to happen is to desolder those wires, get some flux and solder wick, clean all the joints, and properly resolder everything.
Those pictures show really horrible soldering.

[Valen]

Perhaps the right time to consider connectors instead of direct soldering. Like shown in the atto pilot datasheet/guide

[Valen]

Yes, at first I got no contact so I tried the reverse point of C and D and still no good.

The exact opposite side of that C hole (known as a "via") is just as likely covered with a coating. Your best bet for conductivity is the solderpads it goes to and comes from.

[Flydigital]

I have now removed it and cleaned up as best I can.
I'm not sure however if it works properly. I have done continuity test across various points and have found a common connection - see pink spots. Any one of these points connects with all the other spots. Is this correct? I've removed all solder but it still makes contact.

I've looked and the '001M' sandwich and see no bridging inside either.

[Valen]

It is true that there is conductivity between all 4 corners of the shunt resistor. So that is no problem. It is the resistance between them that makes the difference in getting the thing work. 2 of the corners are the power in- and output (in the center of the board). This should have the specified resistance as displayed on the top. (2 1mOhm shunts in parallel= total 0.5 mOhm for the 90A version) The 2 other corners of the shunts (near the plane's image and the array of resistors and capacitors on the other side) are the sense outputs going to the amplifier inputs. These do not carry the full current, but only the voltages of each side of the shuntresistor. The voltage difference between them means you can calculate the current through it based on Ohm's law.

It's impossible to see where the traces are going to under the shunt-resistors and the amplifier, but I think those pink points you measured do indeed have continuity together judging from the schematic in the datasheet on the product page.

Now the definite verdict:
In the top image you can clearly see that one side of the currentsense output connection is gone. As that is also the side that has the trace to the output of the amplifier that trace is definitely broken. So it's either getting a new Atto Pilot current sense board, or bodge solder the current output wire to the tiny resistor capacitor near the amplifier. Make sure you have other ways of strain relief like glue as it will easily rip off when pulling on the wire (or vibrations when flying). Considering that you soldering wasn't that good to begin with this is likely a far fetched idea to try. But you've got nothing to loose I guess.

What sort of soldering wire and iron did you use? Non-leaded? Is the temperature ok?

[Flydigital]

So I may still have a chance!
I think the soldering iron is not hot enough to do quick solid joins with the 10AWG wire. It sucks all the heat away. Its 30W power. The soldering wire is 1mm thick but apart from that it is unlabelled. No idea if it's leaded or what that would mean. I guess I should do some reading on soldering.
So to rescue my current out sensor I will solder to the tiny resistor near the amplifier. Is that point I have marked here?

[Valen]

No, that is the via that connects the traces on both sides of the board. This is far too tiny to solder to. You won't have the surface to solder onto. The resistor and capacitor are a bit further along the trace. The resistor has the code 84C on it (which somehow does not agree with the datasheet) It's ok if you solder the wire to both the resistor and the capacitor next to it (light brownish part), they are connected with a trace on the via side of them.

[Flydigital]

Success!
I got it working and manage the tiny solder onto the spot you said.
I've run it into the telemtry port on my Frsky and now have accurate Amp readings (I compared using an inline amp reader)
Thanks so much for the help you have given.

Clearly very sloppy soldering all over the place and fried connection caused the problem!

[Valen]

Excellent.

[Ross Robotics]

Just noticed this thread. Not sure why I didn't see it before, but have to say nice job Valen and also to Fly.. Threads like this gives me the satisfaction to keep helping people here.

Fly, you can also click the green check mark on one of Valen's posts to mark this issue Solved.