I have two Tarsiers in RacerX design TPU canopy used with whoop style AIO and I don't have any noise issue.

Hi Ronin,

Good diagnostic work        AND ...  your documentation and explanation is SUPER 

You know, I have been using the 26x26 "whoop" boards in my builds. These things have excellent isolation. The "squishy" rubber grommets effectively isolate the board from the 2mm bolts and I always use extra nylon "washers" or spacers.  Yeah, I typically don't cut the bolts. First, I have a full range of 2mm bolt lengths {from 4mm to 30mm} so I just pick the size that fits the best, then use washers or spacers as needed. 

On my Phantom-4X build, I drilled one of the 2mm arm attachment holes a little too big, so I cut a tiny piece of antenna tubing and used it as a sleeve to take up the extra space; it worked perfectly. Now, this method could also be used to isolate the bolt from the frame. 
Just a thought. 

Although I haven't used the Tarsier yet, I do plan on it; just haven't gotten a build for that purpose yet. At this point in time, I don't want to risk losing a 90 dollar video subsystem. It is good to know these issues so when I do get one, I shouldn't have any problems.

The more I think this through this guys, the more my head hurts.  

How can it be related to the Kakute stack when everything in that stack is insulated?  The FC uses rubber grommets and there are no metal rings around the ESC holes.  In this particular Gecko build, I have a nonconductive 20x20 stack plate on top with a TBS Crossfire nano heatshrinked and velcroed to it.

But yet, when I literally unscrew the Tarsier stack screws just enough so the head of the screw was not touching the frame, the video had no noise.  Screw it back in and noise.  Add washers, no noise.  That was literally the steps.

As far as what scotty mentioned regarding the heat sink and issues with GPS, I would add that I have a 2mm TPU spacer elevating the Tarsier stack from the frame.

I think the take-away here if there is one is that if you are having noise issues with your Tarsier, try running some throttle while the stack is unscrewed.  If that removed the noise, add insulation to your screws with a plastic washer and possibly some heatshrink where it would otherwise be exposed to the carbon fiber.

While I did run 4 packs with zero video issues after doing this fix, I have not flown it since then.  I wouldn't put it past this thing to have, for whatever reason, video issues again.  It is still 2020 afterall.  

I couldn't resist but to think about it more.

The only two things I can identify that is different from the Kakute build with no video issues but using the same Tarsier and Kakute stack configurations is the camera and motor wires.

The camera on the Gecko, if say the screws were somehow conducting from the camera, has a clear path to the Tarsier stack screws from the metal camera cage on through the frame. The Kakute camera is held in place with TPU. Could I have screwed in those screws too tight and somehow causing some sort of connection to the camera electronics?

The motor wires on the Gecko are run under the Tarsier stack to the ESC in the rear stack. The motor wires on the Katana are run outside of the frame to the ESC in the rear. Could the bunch of wires under the Tarsier stack create some sort of electromagnetic field interference thats running through the frame and up through the stack screws?

Bah. My head hurts.

I don't envy the issues you're going through but definitely appreciate all the documentation you've done on such. If I get a Tarsier in the future and have issues at least there's some groundwork to try and start sorting issues out.

Thanks darkshot. That's precisely why I go through the trouble of posting it. I know others might be thinking lol wow this guy has some bad luck or something lol. That's ok. So long as others don't suffer as much as me then it's all good.

Thank you for documenting all this Ronin!

Yesterday I also struggled with the same problem while moving all of my Gecko's guts over to a new Micro Apex frame. The analog image from the Tarsier showed lots of interference and diagonal lines when arming and even more when applying some throttle.

After going through pretty much the same frustrating process of checking, re-wiring and replacing components one by one, I also discovered that the issues stopped when the Tarsier board wasn't screwed down. After I found your post I tried to figure out how to isolate the screws but on the Micro Apex it would not be easy without drilling out some holes or perhaps using some flimsy plastic screws.

But then I realized that I didn't have this issue on my Gecko frame, even though it was also screwed down firmly with metal screws. Grounding the entire Tarsier stack also seemed like a good thing instead of a bad thing, it would create some kind of Faraday cage to keep noise out.

So I took my multimeter to check the conductivity of the carbon fiber and see if it was connected to the ESC's or FC ground. It wasnt, like it should be. Then I checked if maybe it was connected to anything else and it turned out that there was a connection between the wires of motor 3 and the frame! I was sure I had insulated the racewire boards well and that my motor screws were not touching the motor windings, but on that particular motor one of the screws had pierced a tiny piece of heat shrink that was supposed to protect the soldering of the motor wire to the windings. I moved that joint out of the way (like on the other motors) and protected it with a bit of glue, and all of my Tarsier interference trouble was gone!

So now I wonder if any of you have the same hidden problem? Could you check the resistance between the frame and all motor wires? Just so you know, I found that the side of the carbon conducts way better than the top or bottom, and I had less than 50 ohm of resistance between the carbon and the motor wires.

(28-Nov-2020, 02:57 PM)jasperfpv Wrote: I have two Tarsiers in RacerX design TPU canopy used with whoop  style AIO and I don't have any  noise issue.

9 months after my statement, I moved  one Tarsier V2/ JHEMCU AIO from a RacerX FT5 frame TPU onto a Fpvcycle Fouride 5" frame. 

Instant jagged horizon lines that increased with throttle.  Went on a merry go round of replacing VTX , bigger caps, rewiring, etc. Days of frustration. Completely forgot about this thread  but thanks to google I ended up back here. 

I removed the metal screws from the Tarsier stack and the lines finally disappeared. Hooray.  I  am using double side tape and plastic M2 screw to mount the Tarsier in the interim. 



LOstB1t,

Good detective work.

I will try to check for conductivity between the frame  and FC/ Motor wires when I can find my multi meter.

Now Jasper, you should have your multimeter with you at ALL times. Please take your responsibilities to your hobby seriously!! Lol…

(15-Aug-2021, 12:34 AM)ph2t Wrote: Now Jasper, you should have your multimeter with you at ALL times.  Please take your responsibilities to your hobby seriously!! Lol…

Badge of shame for this hack amauteur builder.