MambaMax ESC quirk
 

I was goofing off with my Mamba powered jato tonight when I noticed an odd thing. If I go forward, brake, and then go backwards, it works fine. However, if I go backwards, brake gently, and then try to go forward with medium gradual throttle, I get tons of cogging. I have to go to neutral, then add the throttle and it goes fine with no cogging. Also, it does not do this all the time, just once in a while. At first I thought it was just a matter of the ESC not being able to read the rotor position, but how would that be different from when it first starts out? Has anyone else seen this?

For those who don't know; This is on a Jato: spur=54T, pinion=22T, total gear ratio=9.63, 8.4v 2S2P 5.8Ah emoli pack, 4.7lb truck fully loaded.

Is it still going backwards when you pull foreward on the throttle?

Nope.

Have you tried adjusting the startup power at all, Brian? It is odd that it does it intermittently - I will have to try to duplicate the phenomenon on my truck.

I've tried to reduce all the settings on the ESC for the least amount of startup power as possible to tame the spinouts while still giving me decent punch. Below is the output of my saved settings file (it's cool that it outputs plain text rather than compiled code, makes it easier to post):
The throttle curve is actually two linear lines. From 0 to 75 throttle (X axis), the slope is linear to 5.5 lines on the Y axis. Then, from 75 to 100 on the throttle, the line's slope is steeper from 5.5 to 10 on the Y axis. I have no idea why the brake/reverse curve is so complex. I never messed with that one at all.

Update on this.

I haven't had this happen much lately, but I was talking to a guy at my LHS today and asked him why he doesn't carry the CC stuff. Well, he said several people were having trouble with them during racing. If they jammed on the brakes so the tires would lock (stopping the rotor), then try to go again, there would be large amounts of cogging-like action. At first I said I didn't have that problem at all, but then got to thinking about this thread again and remembered that it sounds very familiar after all. During races, I tend to slide around corners more than stop so it makes sense I wouldn't see it very much.

Glassdoctor, he said you had this problem too during races. What are your thoughts? Did you do anything to alleviate it?

If enough people report this, maybe CC will release some type of firmware revision...

On a rare occasion i would get a slight hesitation after hitting the brakes, but i rarely ever fully lock the brakes

If you have it programmed from forward to reverse, the truck all of a sudden brakes (well brake, it's more of going into reverse) hard and throws itself in reverse, this is the same with the mamba 25.. Perhaps the guys from your LHS missed this setting..?

What is your forward to reverse setting?

my mamba doesn't like to start from a stand still... i've tried all the different combos of start power but i haven't been able to resolve it yet... i'm hesitant to contact castle as i'm using it on a g2r revo on 4s with a plettenberg maximum w/fan... gearing is 16/51... i think the fact that the motor is 4 pole has something to do with it... once it gets rolling, i mean even just for a couple of inches, its silky smooth... but from a stand still it just cogs... i found that if i give it full throttle it actually doesn't cog... only problem is if there is any sort of grip, instant backflip... lol... btw i've got a fan on the esc and it doesn't even get warm after 15 min of hard running in a parking lot... the motor feels warm to the touch i would get around 80-90 F and batts are cool... i'm using apogee limn 5800's...

those plettenberg have got a nasty startup.. What you can do is set your gas-curve to start at 5 percent or something, so it's starting a bit more aggressive. Did you tried changing the timing?

and did you programmed it to your setup?

i will try the curve idea... i played around with all the timingh setting along with the startup setting and they didn't seem to make a difference... once i've tried the changing the curve i will post the setup file

I hesitate to get into this too much because there are a lot of unknowns and it's difficult to get on the same page... what I "see" you might "see" a little different.

First off I can say I have talked to Castle about some issue I have noticed... I don't know what can and will be done to address certain things. Some things they (Castle) don't "see" so it's hard to say what they can do about it.

The cogging from a dead stop is a problem with the big cars/big motors. I'm sure certain motors do this more than others. I have run my 1/8 and barely saw this at all and other times it happened a lot and was a big problem.

If you ever have this cogging on initial startup definitely don't keep on the throttle. Let off immediately and try giving just a little throttle to get the motor to start. If you just jam on the throttle when it's sitting there stuttering it basicly is shorting out the esc and it will get very hot very fast... wiring and all.

I actually smoked my datalogger one day when I was getting cogging and decided to see what would happen if I stayed in the throttle a lot etc... bad experiment. The wires got hot.. the plugs on the eagle tree actually desoldered. The MM still works fine btw... just had to let it cool off.

Did you programmed it to your receiver/transmitter?

and what is your forward/reverse setting?

Another thing; the plettenberg is a rather stiff 4 poler.. The stator 'grabs' the rotor pretty hard. Most controllers don't like the plettenbergs because of this.

You can feel a large resistance when turning the shaft by hand. (for those who know the Neu's and who turned it's shaft; it's about 5 times more resitance to turn a plettenberg)

The glitch BrianG describes doesn't sound like the same issue I have seen.

Here's the issue we are having while racing:

This is in 1/10 offroad buggy, 4wd, and truck, running good 6 cell packs, forward only or reverse lockout, two different high end radio systems, five different cars and rx's, differnet electronics layouts, two different drivers, etc etc etc

It appears as if the MM can "misfire" in certain situations, when there is just a small amount of throttle input. What happens to us is when coming through a sweeper we let off the throttle and let the car coast a bit (with the drag brake activated) and then gently roll back on throttle.... sometimes the car will "glitch": the rear wheels lock up just for a split second and then back on power. We belive what is happening is that the MM gets a misreading on the rotor position/direction and it "fires" backwards once... then immediately goes back to the correct sequence. It is just enough to make the car loop out sideways... just like a quick stab of the brakes in the middle of a turn.

This happens to me about once per race on average... sometimes I have seen it several times in one race. I didn't notice the problem until I started racing on the smooth indoor tracks recently, where this behavior is obviously not just a bump or rut in the track messing with the car.....

We have the brakes dialed way out... the brakes are not strong enough to even make the car do what it does. And we have the drag brake set to around 30%... pretty strong... so we only need a little more brake available when we want it.

Setting the exponential on the radio to "soften" initial throttle response seems to make the problem worse, as well as setting the drake brag higher may make it happen more.

Castle's response so far is that they are blaming the connectors... they said that one guy reported a similar issue and that the problem went away when he direct soldered the motor. They think possibly the very small low voltage pulses that the MM uses to get info from the motor may not "like" the connectors. I tried putting a little solder "jumper" on my connectors to see if that would help and my buddy replaced his connectors with Corally style gold plugs that are similar to the 4mm stock ones but are a different (we think superior) design and quality. But we still see the issue.

This week I plan to direct solder my motor and see what happens.

I don't have the problem when taking off from a standstill, it's only when I used the brake to come to complete stop (no motor movement at all) and then try to go forward again. And even then, it only happens sometimes. I have been able to come up with a specific action where it is perfectly reproducable.

Since I'm using the MM in a small, lightweight truck, it recovers fairly well. However, using it in a relatively heavy buggy or MT exacerbates the problem since the ESC will have a harder time to get the truck to move due to load and motor size.

Personally, I think it's just a matter of the motor position sensing algorithm. If CC could recreate it, I have confidence that they could release an update to solve it.

Believe me, I'm all about sensorless designs, but it's issues like this that give sensorless a bad name. It has to be something specific to the MM, because my Quark doesn't do this at all (I tried it).

Glass;

What setting on forward/reverse do you use? forward only? i find the forward/reverse setting a horror.. It blocks the wheels entirely.

Just to make it clear... I get zero cogging in 1/10... and I have not noticed the "racing glitch" with the 1/8 stuff...

forward only... or reverse lockout...(happens with both) I have never run the forward/reverse setting...

I run a M8/Spektrum and my buggy runs a KO Helios/Spektrum

I have tried a lot of different esc settings but nothing "fixed" it... I think Castle needs to look at the software to build in something that makes this glitch impossible. I have not seen it on other controllers....

This is the first software release on the mamba-maxx. I remember the mamba 25 cogging like crazy in the beginning with the higher KV motors. A software update fixed this problem. However, my 8K and the mamba 25, misfired from time to time.. All of a sudden you heard this funky squeak coming from the motor, without moving the vehicle like it supposed to do.. after the software update i didn't had the issue anymore.

When we first talked to them about this they were wanting to replicate it so they could fix it in software... but now it seems they are using the connector as a scapegoat and don't act interested in pursuing a fix other than to say cut off the connectors.

I think they can and will fix it, but they seem to say it's just us and they don't see a problem. Which has me a little disappointed in them right now.

I have some more testing to do... and I will likely visit them at their shop sometime soon when I think I can put the car down and show them so they can see it with their own eyes.

glass; did you called them?

if a direct soldering doesn't work, you've got another reason to contact them?

seems to me like they where only trying to make a suggestion on a for them unknown problem.

Hmm, so it's not specifically related to using reverse then. Well, at least that's one possible cause we can eliminate.

I would highly doubt it's the connectors. Even when the wires are warm after an hour or so of constant use, the connectors are still relatively cool, so I know there is good contact there. And the actual motor wires are MUCH shorter than they were when new - only about 2.5"-3" total from ESC to the motor now.

I really hope it can be fixed with an update.

Does CC have a local office or something?

It's not that the connectors can't handle the load... it's that they think possibly the micro pulses used to "see" the motor doesn't like the connectors being there.

serum... yes both myself and another local racer have talked to them quite a bit about it. It sounded pretty good for a while that they really wanted info and wanted to nail the issue... they were not happy to hear there was such an unkonw bug in the system still.

But once they recalled the guy who said he stopped his problems by direct soldering, it seems they called off the dogs and are content to just blame the connectors and say that they can't replicate the glitch so it's just me and my buddy that has a problem.

The guy who reported the direct solder fix is the 160mph speed run guy (nick?).... the funny thing is that he wasn't even running mambas in his cars.

I haven't called to ask Castle if they know exactly what setup he was running when he "fixed" his glitch. But his speed cars ran the airplane controller and a neu motor.

I have talked to several guys at CC and I will continue to... but the past few weeks I'm kinda taking a break, and haven't done much with r/c except for one weeknight race...

I plan to mess with stuff and see if I can make it do it a lot... so I can go to Castle and show them exactly what it's doing in person.

This isn't over by any means. :)

Castle is in the Kansas City area.... I have relatives in the area so I get down there once in a while.

edit for spelling :)

they are usually really good with customer support... who did you talk to? ask for Joe Ford... i've dealt with him in person before an he's very helpful...

Nick Case was the one who made the 160mph speed run and he was using the CC Phoenix HV-85 controller as far as I know.

Brian, glass; what motors are you guys using? I thought you are using a Neu, Brian?

Glass, what motor was on the racer? was that the original mamba motor?

I'm using the bundled 4600 motor on 2s2p. I tried lowering the gearing too and that didn't help (or if it did, it was not noticeable).

I have talked to Joe, Shawn, and Bernie. All have been good guys to talk to.

We are racing 1/10 offroad with the stock MambaMax systems, motor and esc.

My cars:
B4 with 4600
T4 with 5700
XX4 with 7700

Other guy:
B4/XXXcr with 5700
T4 with 5700

Yep... Nick was running the HV and a Neu 1509 on kokam lipos I believe.

I'm not coming down on CC for this... I'm trying to help them figure out what's going on with the system. I don't think it's just ours that are doing it. It's a subtle problem especially for those not racing. And no offense, but also maybe not so noticeable to those who are not very good drivers. :) And like I said before... on a rough track, in grass,... this would be hard to detect.

And there's a lot we don't know...

It where flightpowers, those are the speedrecord cells.. :p

And 8s of them, too?
Nitro Tc3 uses 8s= 130+mph
The dirt oval rig uses 8s, too. = 160mph

My mm also does this. It's in a touring car, hard-wired running a fiegao 8s on 6 cells. I noticed it on my first run.

Wow! I thought I was the only one when I started the thread since the few replies I got led to dead-ends, but it seems like this is quite a widespread issue after all.

Yeah, the problem is there. Not a huge problem, but they definitely should not be ignoring the issue. Maybe not a lot of people have not reported the problem. I noticed it as soon as I ran it but I just took it as you put it, a "quirk".

Serum, I'm thinking the flightpowers and the kokams are the same cells? I've been told that more or less but don't recall who or where. What's interesting is that we hear raving reports about both being the greatest cells ever made. :)

lydiasdad: Which "glitch" do you have? There are at least 3 different issues mentioned in the thread...

1. BrianG's glitch: cogging when going from reverse to forward

2. Glass's glitch: hiccup when gently coming back on throttle while the car is still moving

3. 1/8 vehicle: basic cogging at start up.

IMO, #2 is by far the most critical issue and the biggest problem, because it has a big impact in using the system for racing.

#1,3 are issues and it would be nice to see fixed, but they don't prevent anyone from using the system and it's not a big problem even for racing. I guess #3 can be a problem if you crash a lot and have to get going from a standstill several times a race... but then this is using the MM out of spec and with aftermarket motors. I think CC will improve on #3 with future software releases.

#1 and #2 I don't know.... I hope they take the time to work out these...

I'm not exactly sure. I only have the problem when the vehicle is at a standstill. So I guess it's problem 1. Mine also does it intermittently as BrianG described. I actually have not payed much attention to it. I will run mine to see specifically when it does it.

GD: While going from reverse to forward (#3) doesn't apply to racing, I think the cause is the same as the other two scenarios. It seems to be a matter of rotor position sensing and how the MM software deals with it. I haven't had the problem where it coggs coming out of a turn, but as you said, that might be because I'm not a skilled driver and simply don't notice it. Or it might be more prevalent on a heavier vehicle where the ESC is running closer to it's max rated specs.

If anyone has noticed glitch #2.... post what setup you have... more info the better.

I gotta think the on road guys would notice this...

I agree it's the same basic flaw in all the glitches... esc firing the wrong coil... backfire of sorts, but there is a big difference between #1,3 and #2.

#1,3 happen when the car/motor is stopped, or at least when it's changing direction. This is a rather common issue with sensorless systems, an inherent flaw.

#2 happens while the car and motor are moving... not just barely moving either.

It is "supposed" to be impossible for this to happen. In fact the first thing the guys at CC said was basicly "impossible... that can't happen....".

But it does happen and we (they) need to find out why.