25-Jul-2016, 02:51 AM (This post was last modified: 03-Dec-2017, 11:11 AM by Oscar.)
Many still are unsure whether or not they should connect the ground next to ESC signal, so here is why you SHOULD!
The main reason for the signal ground is for noise reduction. As you might be aware that many ESC's come with signal and ground wires twisted, because it shields the EMI interference - noise, that could affect the signal wire. Any wire could act as an antenna, the longer the signal wire, the more likely it will get affected by noise and interference. If noise inhibits the signal wire without the ground twisted around it then this may cause problems.
Another source of noise is from ESC power. Because multirotors' motors are constantly changing speed, the voltage across the ground level of a quadcopter can change significantly, therefore it makes a LOT of noise. This can be enough to corrupt ESC signal outputs in some cases.
Some FC doesn't have ESC signal ground, such as the FuriousFPV Piko BLX, in which case you can just connect the signal ground to the the power ground pad.
Let us know if you have any comments
Update: For smaller and less powerful build (e.g. a 3" micro that draws less than 15A of current each ESC), we might be able to get away without signal ground. If the signal wires are not very long, you might also be able to get away from using signal ground, because when it's shorter, it becomes less effective as an antenna Update (Apr/2017) - For ESC that are using DShot, signal ground is strongly recommended regardless of current draw.
Twisted pair actually only makes a significant difference if you're using a balanced signal line and ideally a hardware transceiver, neither of which are happening on our FC. That being said, it will make some small difference, basically acting as a limited shielding. At the very least it won't hurt
Hey Oscar....just finished hooking up my ESC on the new build I'm working on and I did indeed run the grounds as well. Your explanation confirmed what I suspected. So thanks for that!!
25-Jul-2016, 03:58 AM (This post was last modified: 25-Jul-2016, 04:04 AM by FPVXX. Edit Reason: Dose, Dse )
Thanks . I was about to risk it with a ESC self grounding. Now I wont. Thanks. Hi , Oscar , you look like you may be able to answer this question. Dose the LiPo act like a capacitor ? . What I mean is , If I bring the VTX wires as close to the battery , and as far away from the ESC as I can, will this reduce interference in the VTX ? . Thanks for your considerations .
(25-Jul-2016, 03:58 AM)FPVXX Wrote: Thanks . I was about to risk it with a ESC self grounding. Now I wont. Thanks. Hi , Oscar , you look like you may be able to answer this question. Dose the LiPo act like a capacitor ? . What I mean is , If I bring the VTX wires as close to the battery , and as far away from the ESC as I can, will this reduce interference in the VTX ? . Thanks for your considerations .
I am not sure if Lipo acts lke a capacitor, but it would help for VTX to stay away from ESC. Also keep wires as short as possible.
25-Jul-2016, 10:17 AM (This post was last modified: 25-Jul-2016, 01:18 PM by Oscar. Edit Reason: added image )
Quote:Some FC doesn't have ESC signal ground, such as the FuriousFPV Piko BLX, in which case you can use the power ground.
I was thinking about getting one of these for a micro build. Do you mean that I should wire each ESC's signal ground to the FC's power ground?
What about twisting the ESC's power ground around the ESC's signal wire instead? (I just watched a video where someone did that on these ESC's because no signal ground was provided from the ESC.)
25-Jul-2016, 01:21 PM (This post was last modified: 25-Jul-2016, 01:21 PM by Oscar.)
(25-Jul-2016, 10:17 AM)sloscotty Wrote: I was thinking about getting one of these for a micro build. Do you mean that I should wire each ESC's signal ground to the FC's power ground?
What about twisting the ESC's power ground around the ESC's signal wire instead? (I just watched a video where someone did that on these ESC's because no signal ground was provided from the ESC.)
I would do it whenever possible, but for a tiny micro build, you may probably get away without signal ground, because the signal wire is generally shorter. The ESC's also don't draw that much current, so the power might not be as noisy as some other more powerful 5" builds...
(25-Jul-2016, 10:17 AM)sloscotty Wrote: I was thinking about getting one of these for a micro build. Do you mean that I should wire each ESC's signal ground to the FC's power ground?
What about twisting the ESC's power ground around the ESC's signal wire instead? (I just watched a video where someone did that on these ESC's because no signal ground was provided from the ESC.)
It does raise the question of whether we're adding noise by not having sufficient filtering or by adding ground loops, because we don't use isolated DC power source for the FC and ESC's and most ESC's share the same ground path we fall into the trap of a possible ground loop (which can create electrical noise)
What does this mean... Well maybe we should be more aware of ground loops if you think you have a noisy copter. The ESC signal ground wire should be connected only on one side so it can still protect from the noise generated by the back EMF but also leaving out a possibility of ground loop noise.
Let's see if this works. There's 3 pics on the attachment. Top is without a ground wire, bottom left is what the FC puts out and bottom right is the noise from running long servo cables.
(25-Aug-2016, 02:04 PM)Rook Wrote: Let's see if this works. There's 3 pics on the attachment. Top is without a ground wire, bottom left is what the FC puts out and bottom right is the noise from running long servo cables.
(25-Apr-2017, 09:58 PM)DALE Wrote: Hello! How about if you run DShot nowadays - is the signal gnd still relevant?
I would say even more so because some interference that on analog could cause the quad to misbehave may actually cause it to enter failsafe on digital. And the higher the frequency the more critical it becomes so DShot600 is more sensitive to this than say DShot300.
Not sure if the implementation has changed since the time when I was following DShot's development closely but it used to be that if a certain number of "packets" were corrupted in a row it would enter FS.
25-Apr-2017, 10:47 PM (This post was last modified: 25-Apr-2017, 10:48 PM by DALE.)
(25-Apr-2017, 10:11 PM)tozes Wrote: I would say even more so because some interference that on analog could cause the quad to misbehave may actually cause it to enter failsafe on digital. And the higher the frequency the more critical it becomes so DShot600 is more sensitive to this than say DShot300.
Not sure if the implementation has changed since the time when I was following DShot's development closely but it used to be that if a certain number of "packets" were corrupted in a row it would enter FS.
Very interesting... I did not think about it this way... Thx - I will have to find some more info about this.
(25-Apr-2017, 10:47 PM)DALE Wrote: Very interesting... I did not think about it this way... Thx - I will have to find some more info about this.
The DShot signal still consists of pulses, but pulses that have a much shorter length than a PWM signal. The signal ground is still required and is actually even more important as a filtering capacitor has been removed from DShot compatible ESCs to make it possible to even send the DShot signal.
The following 1 user Likes unseen's post:1 user Likes unseen's post • Oscar