Castle Mamba XL2 ESC cogging
 

I'm using a newly bought Castle XL2 ESC with the following setup:

- Car: Carson Dirt Attack XXL (Redcat Rampage XBE clone) wheighing 9.3Kg (20lbs 8oz)
- Motor: Turnigy TrackStar 1/5th Sensorless Brushless Motor 730KV (Size: 5592)
- Batteries: 2 4s Zippy Flightmax 8000 30C batteries wired in series for 8s

XL2 settings:
- Firmware 1.32 (Date 4/5/2011 -> is this the most recent one? ... castle link says yes ... i wondered about this since that's more then 2 years ago)
- Startup Power is set to low
- No punch or torque control
- lipo cut off is set to 3.3v

No additional cap bank and i'm using the BEC to power the receiver and servo.

During the first few minutes startup is as smooth as i could wish for, but with time it degrades more and more up to the point where the motor coggs badly and the car almost isn't able to start on it's own. This only happens if the car comes to a sometimes nearly or complete standing still and then if i try to get it going again = cogging.

Two fellow members on our austrian forum (german speaking) have reported the same behaviour and haven't found a solution to this.

Anyone else experiencing the same issue?

Any firmware update planned soon or is there a firmware i can switch back to helping with this issue?

Anything else i can do to ge a constant smooth start up up to the moment the ESC hits LVC?

What i noticed is that when startup works as it should one can hear a dedicated sound from the ESC the first second as the motor starts up from 0 rpm. Can't hear that anymore when cogging at startup begins. As if the startup sequence isn't engaged any more after a few minutes.

Pic of my setup:
http://www.rc-monster.com/forum/atta...1&d=1376632743

IMO

- Start power should be med at least - The inertia on this thing will be big.
- Re-check gear mesh.
- Get a BEC Pro - It adds a couple of wires but your ESC and servo will be much happier - I'm sure you must be using a digital 1/4 size servo... PS - Is the monster Fan off the ESC too! Pretty please get a bec :)
Extra idea - recalibrate / check radio/rx - Are you getting central drift so that it doesn't realize its at nuetral

I will add an external BEC for sure. The 5A BEC seems a bit weak for my taste either. The 12V 80x80 fan is powered off one of the 4s batteries. Gear mesh is set relatively loose compared to my 1/8th cars ... smooth running with no binding. Start Power low was the default setting ... i thought this will be easier for the batteries so i kept this setting. I don't think that there is a neutral shift, as i'm using the same radio with different cars and never have experienced troubles when starting from 0 rpm either back or forward. Thanks for your inputs and ideas.

There is one member on our austrian forum who has reported that on his race track 9 out of 9 XL2 ESC are showing the same behaviour. When changing the ESC to an MGM or Atos the behaviour is gone. So must be the ESC. While racing the issue is sort of non existent since they won't brake their car to 0 rpm.

On the other hand we have one member with a Baja 5e who is not experiencing this phenomenon with his blur rtr setup and another one who is using 2 XL2, one in his losi five and one in his Crick Corr Truck and he is reporting having the same issues as i do.

I haven't searched other forums but this can't be a problem which happens only with my setup.

Video of today illustrating the constant degradation of the cars start up from 0 rpm (130mb).

Runtime almost 10 minutes. 6200mAh used out of 7000mAh usable.The car is geared for 44 mph (70km/h).

I pulled over a few times during the run to check the startup behaviour. I think it is noticeable that over time start up becomes harder and harder and cogging gets worse and worser. Especialy at the end of the video one can see that cogging is at a point you don't want to have it.

Hi Othello,

I have asked my engineers to take a look at this problem and see if they can reproduce the issue.

Please give us a couple weeks to see if this is something that may be fixed by a software patch.

Thanx!

Patrick

Thanx Patrick. Other than the described issue the XL2 ESC is performing as i have hoped for. I will check for firmware updates now and then.

Any news or findings on this yet, Patrick?

I have similar behavior on my Baja with an XL2 as well. I'm going to be adding a cap pack and an external BEC this weekend to see if it clears it up. It's not quite as bad as this one shows, but it's very noticeable, especially about 5-6 minutes into a run.

I will update with regard to improvement after the change.

I should have posted my XL2 settings earlier as i'm now driving an xl2 with castle 2028 powering a much heavier Losi 5iveT without any cogging issues (and it also worked with the trackstar motor powering the car shown in the video posted above).

You may want to try these settings:
http://bt2007.braintrust.at/ds/rc/lo...2_settings.gif

On my Mamba Monsters I could also observe cogging that looks and feels the same as the XL2 cogging, although not as bad. It seems it is connected to ESC temperature somehow. After I removed the case from my ESCs and put a bigger fan on them, the cogging was reduced significantly.


Here's what I found out about the XL2:
The two parts of the cases are glued to the PCB with thermal glue that doesn't hold on too strong. With the ESC that I got, the bottom half almost fell off on it's own and there was dirt on the thermal glue, so it wasn't really sticking anymore. I have seen the same on rcs2022's pics of an opened XL2 here on the forum, one could see dirt there also, so it must have seperated before. The plastic case around the heatsink case holds the two parts together, but it doesn't apply pressure which is needed to keep everything tightly together for good heat conductivity between the PCB and the heatsink.

Also, the fan cannot be very effective, the grill on it doesn't allow for good airflow and the upside of the ESC casing also doesn't allow for good airflow.

You could try the following to fix the cooling:
1. Remove the plastic casing and clamp the two case halves together using some threaded rods in combination with braces.

2. Use a big (60x60 or 80x80) high-flow fan and mount it half an inch or so higher (not directly on top of the ESC case).

Thank you lizard for sharing your findings. This corelates well with the fact that while the XL2 logs do show temperatures up to 100°C the aluminium cases remain relatively cool (around 45°C max).

Would be interesting to see how much of a difference that makes when the case is clamped together firmly.

Problem is also, that there are FETs inside which have no direct contact to the case, just an indirect cooling path through the PCB.

If somebody tries that, it might also be a good idea to relocate the two capacitors that are inside the case (opposite to the two caps where the battery wires are) to the outside. They make direct contact with the FETs, so they heat up the FETs and vice versa, not good.


Edit:

Pictures of rcs2022 XL2:
http://www.rc-monster.com/forum/showthread.php?t=31997

Pictures of Revorocks XL2:
http://www.rc-monster.com/forum/showthread.php?t=31411

Pictures of my XL2:
http://s54.photobucket.com/user/liza...06/library/XL2

Hi Othello, I know it's a bit of an old thread, but I thought I'd try my luck:

How do the settings you described to run successfully in the posted pics, differ from the settings that got you the cogging problems? Only difference I see in your posts would be a change of the LVC, from 3.3 to 3.5 V/cell, no? Or am I overlooking something?

Did you end up using additional caps and an external BEC? Did that reduce ESC temps and were those additions a factor in the solution to the cogging problems you think?

Thanks!

Edit:
Lizard was so kind to link me to your thread at offroad-cult and I read about your experience with a failing XL2 while using Castle Link, seemingly because there was a load (external BEC and fans) connected to the battery leads while connecting the ESC to the link. Seems your new XL2 has been running well right? So the solution would be to put a switch between any loads (fans, UBEC, other) connected to the battery leads, correct?

What seemed to have worked was (i think) more due to the reason that i now usually make a short brake after 5 minutes into the run (which helps the ESC to cool down inbetween).

I you take a look at the following video (FG onroader with XL2 ESC) ... you will notice heavy cogging right at the end of the run when i tried to bring the car in. The esc had logged around 100°C at that point. At the beginning of the run there is absolutely no cogging. And i think i used the settings i have posted above. It really depends on different factors how bad it might cog. Can't remember i have seen the Losi 5T cog so bad after a longer run.

Yes: additional cap pack, external bec and dual 5.5mm plugs. I'm also using a 10ohm resistor when plugging the batteries to the esc (ESC is set to 8s ... not auto lipo). I think those "goodies" help the ESC to survive and work reliable. I also have installed a plug to disconnect the UBEC from the ESC when i'm retrieving data via castlelink (as it would be otherwise powered as soon as you connect castlelink to your laptop/PC).

With all this small goodies i'm now runing two XL2 esc without issues.

Would be interesting to know when the "xle50" castle car esc (12s?) will hit the shelves.