Turnigy TrackStar 18A ESC

[rossrov]

KR2: No problems with entering program mode. Have to remember to wait until the second set of descending musical "notes" though before you go from full throttle to stop. I watch the light colour when doing the throttle calibration - though the instructions say you only need to do this bit once. Also have to remove power supply completely (switch off at battery) before turning on ESC arming switch for programming.
Tried any slow starts yet? My only issue with the ESCs is unable to start consistently in say under half a second.
Biker Bones: How do your ESCs start up please? Any sort of delay or hesitation, particularly when commanding less than 1/2 throttle?

Ross

[bikerbones1968]

No issues at all. They start up soft and slow as they should.

[rossrov]

Thanks Biker Bones. Good video. BTW, liked your router jig for doing O-ring grooves over in the tools section.

[KR2_Diving]

Also have to remove power supply completely (switch off at battery) before turning on ESC arming switch for programming.

Eureka!!

Thanks Ross! That was the ticket! I was not removing the battery power, just relying at the ARM switch!

I also noted that the motor MUST be connected to enter the programming mode.

Here is my procedure for entering programming mode:
1.) Ensure motor is connected to ESC.
2.) Plug battery into ESC
3.) Set SERVO out on ARDUINO to 180. (FULL THROTTLE)
4.) Turn on ARM switch of ESC.
5.) ESC flashes LED and rings once.
6.) Wait a few seconds.
7.) ESC flashes LED and rings 4 times indicating it is ready for CALIBRATION MODE.
8.) Continue holding FULL THROTTLE.
9.) ESC led flashes while beeping.
10.) Wait a bit more...
11.) ESC flashes LED and rings 4 times.
12.) Wait a few more seconds.
13.) Wait until ESC is beeping in time with red LED. (beep---beep---beep---beep---)
14.) Set SERVO out on ARDUINO to 90. (NEUTRAL)

ESC is now indicating you are at question 1, and asking if setting 1 is correct.

To answer YES: Set SERVO out on ARDUINO to 180. (Full Throttle)
-ESC then goes to next question prompt.

To answer NO: Set SERVO out on ARDUINO to 0. (Full Brake)
-ESC then goes to next setting option for active question.

Now I think I've got it!! Thanks Ross for the hint!

[KR2_Diving]

Ok, time for another weekend update! (After spending last night in the Emergency room with a friend, I thought it best to crack down and focus on the ROV for the time being!)

I am having a bit of an issue with the slow starts on the TrackStar ESC... I have tried playing with a few of the settings, but I can't seem to find the silver bullet solution...
You can see the video below:


I have set up two ESC to compare the performance of different settings
(unless noted, setting is the same)

ESC 1 ESC 2
1.) Brake/Reverse Type: 3 (Forward/Brake/Reverse)
2.) Brake Amount: 4 (100%)
3.) Reverse Amount: 4 (100%)
4.) Punch Control: 2 (Medium) 5 (Disabled)
5.) Drag Brake: 5 (40%) 1 (Disable)
6.) Throttle Dead Band: 2 (normal)
7.) Voltage Cutoff: 2 (Auto-Lipo)
8.) Motor Timing: 2 (normal)

In the attached video, ESC 1 controls the motor that is closest to the top of the screen. ESC 2 controls the other one...

On the Display, you can see the value that is output using the SERVO.h library. I control this via a Pot that I have connected in. (video is not the best... but it does the trick, I hope...)

Any suggestions on how to clean up the soft starting based on settings on the ESC?

[rossrov]

KR2: The ESC behaviour you're finding is exactly what I have been dealing with. (had a minor panic attack on hearing the beep-beeps and start-up noises in the video!). We are being too gentle with the start-ups. I think you will find that by altering your code to initially send a greater value for forward starts (and lesser for reverse starts) the start-up times will be minimal. In other words, when the pot goes from 90 (neutral) to anything above that, what you actually send to the ESC will be a burst of 110 to get the motor moving, then reducing or increasing this back to whatever the true pot value is, by which time the motor will be moving and responsive. So stick a few lines in the code between where the pot is read and where it's variable is sent to the rest of the code that creates the PWM.

Like bringing the revs up before letting the clutch out when driving a car from standstill, but different.

To test the above suggestion, alter your reverse-neutral-forward button code you have already written for programming/calibration, but instead use it to control the motor speed. Try reverse=70 neutral=90 forward =110 and see how that goes

I tried mucking around with the brake settings etc but did not find it helped much, if at all. In fact, MAY be better to have no brake so motor is still moving should you quickly change direction

A big part of my particular ESC start issue was the PWM period seemed too short, despite measuring it as what it "should" be. Re-loading my original test code which had a longer period made this part of the start problem practically disappear.

[rossrov]

DNA and KR2: How you going with the ESCs? Possibly of interest I have posted some pics of my ESC and battery/electronics mounts which also function as a heatsink, should heatsinking be required. First I was just going to use a big washer on the outside but then found a lump of brass in the shed, so drilled and tapped a hole in that instead. This was handy to cable tie the wires to while potting, though that's getting off topic. Ideally all the metal parts would be aluminium or copper. Since taking the pics, I have fitted the ESCs. Made a point of checking that the ESCs own heatsink fins were not internally connected to the ESC wiring.

[dna1990]

We did get the parts in finally from HobbyKing, sort of a hassle thru our local post office...but anyway.

The kids were able to hook up one on a test bench and have a trim pot on the Arduino that will run the motor thru the ESC, variable speed and reverse. So big steps. But that is far as it went and they have all started summer break and have scattered to various camps and other family activities. So no heat testing or multi-motor control yet. In fact, have not put any load on the motor, still just bare shaft...so much left to test in that regard.

The ESCs are MUCH smaller than I had envisioned.


But your photos don't show the (pre potting) detail as to where the wires are in relation to the brass sink. Meaning is there one big hole and the brass and wires are stuffed in a potted? Or the wires have their own small holes, but in close proximity to enable a single potting glob? If the wires right next to the brass (or other metal sink), could there be any expansion/contraction that could create a small void between the wire and the metal (where potting has not reached). Or could the sink heat up enough to damage the wire insulation. I doubt the later, but curious.

Oh wait, I just re-studied. The brass is only what we see I the photo, correct? Meaning the hole diameter is not the size of that brass but rather the bolt from which is connected underneath. Which would mean the wires have their own small holes....yea?

[rossrov]

Thanks for feedback DNA. Good to hear the bits turned up and you've had a play. Yep the ESCs seem very small for the amount of current they are claimed to do. In terms of power, that's about the legal limit for no registration and no licence needed assisted bikes and scooters in this country, so unless one's ROV is running flat out probably no need for concern.
Good points you made, and yes, just a small hole for the stud (chopped-off bolt). If a more effective heatsink was required, you would use something other than 8 millimetre stainless, stainless steel not being the best conductor of heat.
I thought about the expansion/strength issue, but even at worst case, epoxy is pretty tough and somewhat flexible.
(Off topic but anyway..)The wires are solid copper (=not stranded) with the insulation left on to help prevent them breaking off where they go into the potting and of course to prevent shorting out on the brass. The insulation has been removed 50mm or 2" from the end then the wire bent into an "S" shape with the long straight bit on the end which goes through a small hole drilled in the PVC cap. Cable tie is slid down over the bunch of wires and the brass. Fiddly! The grey stuff on the inside of the cap around the wires is polyurethane sealant- this stops the epoxy running out. Masking tape is wrapped around the circumference of the PVC cap to form a dam then epoxy poured. I used some local stuff similar to your WEST or System3. (fibreglass laminating resin as used in boatbuilding etc). If using epoxy, follow safety recommendations. Google "epoxy hazards". Some people have allergic reactions to even small amounts. Clean and sand surfaces to be bonded.

[jpWA]

I'm confused. do these hobby ESCs work with brushed motors or just brushless?