Inside the Mamba Max

[DKNguyen]

Hello. Does anybody have pictures of inside the Mamba Max? Or was it the Mamba that everyone uses?

[BrandonW]

the mamba is for 1/18th applications like tlt's and the mamba max is alot bigger for e-maxxes and such:

http://www.castlecreations.com/products/mamba.html

[WheelChair]

Why do you want a pic?

The circuitboards are encased in epoxy...

There is basicly nothing in the plastic base....

[DKNguyen]

Ah, that's too bad. I was wondering (among other things) what was under the heatsink. I'm having trouble figuring out how to set up a small enough heatsink on the PCB while keeping everything small (like stackign PCBs, etc).

[WheelChair]

Originally Posted by DKNguyen

Ah, that's too bad. I was wondering (among other things) what was under the heatsink. I'm having trouble figuring out how to set up a small enough heatsink on the PCB while keeping everything small (like stackign PCBs, etc).

My guess is they built it that way to make room for the USB port.

Also, if you are thinking of space saving.... Remember, things like radio's and speed controls need space so that they don't get feedback and signals from something else...

Some Electronic equipment needs to be sheilded from interfearance.

I took mine apart to see if it needed any further waterproofing. The USB port is the only thing I am worried about there...

I would leave it in the case... its not that much space anyways....

[DKNguyen]

I was asking to see how I should design my home built one so I can mount it on the Clod axle (opposite of the other axle that holds the steering servo). PLenty of space up there away from everything else. DOesn't really matter if it needs shielding either...just spray the outside with copper spray.

I really hate stacking PCBs see.

[Big Mike]

You are home-building your own esc?
This I gotta see...

[DKNguyen]

Originally Posted by Big Mike

You are home-building your own esc?
This I gotta see...

Well, yeah. I'm not a rock crawler. The only reason I'm asking here is because I'm using a rock crawler chassis as a platform.

[5150bronco]

well, post some pics and tech info if you do, that would be nice....

[DKNguyen]

My two biggest problems right now are how to pass that much current on the PCB (not so much a problem as it is expensive as they cost quite a bit to get made in small quantities, especially for one with 6x the regular amount of copper). The second is I kind of want some pretty ridicuolous switch speeds for the transistors but also want to get a single IC to drive all 6 transistors at the same time rather than using 3 separate dual driving ICs. Not really a problem either...just an annoyance.

[Big Mike]

Well, yeah. I'm not a rock crawler.

Does'nt matter to me, dude. You're into RC, so what if you don't crawl?
There's definitely many other ways to have fun w/ RC besides crawling.
I did'nt mean to come across as sarcastic or anything - If you can successfully construct your own esc, and have it perform the way you expect, you are one techno-genius guy. You get mad props from me for even trying it.
Rock Crawling taxes an esc probably more than any other genre of RC, and the "ideal" crawling esc does'nt really exist at this point (Well, Mamba Maxx is pretty close) if you could successfully construct a custom esc that has the specs and performance you desire, well, the implications could be huge...

[DKNguyen]

Originally Posted by Big Mike

I did'nt mean to come across as sarcastic or anything - If you can successfully construct your own esc, and have it perform the way you expect, you are one techno-genius guy. You get mad props from me for even trying it.

I ain't paying 6 grand a year for nothing!

I'm not actually into RC at all. I'm just hitching along for the ride. RC truck parts are just so much better and cheaper than what we electricals can build on our own...not to mention pimp.

[Big Mike]

I ain't paying 6 grand a year for nothing!

OK - now you've really got me confused!! (school?) Well, good luck with your project anyway!
Now you've got me curious - what ARE you trying to accomplish? If you'd care to elaborate...

[DKNguyen]

Crap...it logged me out while I was typing so I have to retype all this...

Yeah, university. I'd like to think my money got me something that Ican use for enjoyment. So here's my rant:

Since I started with robots, I've wanted to build a robot that's not limited by where it can go (like everywhere a person can). I use stairs as my benchmark since those are some of the hardest things to climb up and it really limits where a robot can go since stairs are everywhere,

Of course, robots with just two drive wheels that can only travel on flat ground (Segway-type two-wheeled balancers being the exception) were off the table right away since they can't go anywhere except for artificial environments. Segways were neat, but they lack expandibility (you can't exactly stick anyting new you build onto it because of the way it works and the mechanics of the design). They can't handle really rough terrain either. They travel fast though!

I started with Mars rover-type robots in mind. Those robots are totally impractical for hobbiests since they have itty bitty super expensive, heavily geared motors with advanced harmonic gearboxes that are shoved into a giant hollow custom-machined wheel that can contain the entire motor. As if that weren't enough, the motor also has to be able to support all the weight of robot since that's just the way it works. *Deep breath* and not only that each wheel is mounted to it's own little servo steering mechanism that also supports the weight of the whole thing!

Snakes, and legged robots were not very practical either. They can go many places but are slow, complicated, expensive, lack expandibility (it's kind of hard to build new experimental parts to go onto a snake since there are so many mechanical limitations). They are also power hungry, delicate, and heavy.

Tracks were neat. But smaller than life-size, larger than toy-size tracks don't really exist. The few that do are all made of plastic and rubber. They're expensive hard to work with, have a ton of parts that usually have to be customed and yeah...

With what we have, wheels are almost the only way to get good speed, efficiency, and load carrying capability. Most ATR robots you see in hobby are not much more than motors bolted to a metal plate. The lack of suspension and clearance really limits things. So the only thing left was all-terrain wheel vehicles that weren't as complicated as the Mars Rover suspensions, but had to be a lot more robust than merely some motors bolted to a metal plate. Of course, I'm talking about suspension but that's pretty complicated stuff that can get expensive- except in RC. RC is about the only place you can find mechanical stuff like that at good prices.

So for the longest time I was planning to rip the suspension off of HPI Savages and bolt 6 of them to a metal plate to get a 6x6 all-terrain rover. Each wheel would have it's own steering, and direction control to get skid-free zero-radius turning. But I wanted all metal parts too and we all know those cost a fortune especially if you need to get every part separately. It would have been expensive and a lot of work to adapt that all.

So I wandered the net and came across "rock crawlers". I didn't even know such a thing existed. Later I found out there were actually full-sized versions that people actually rode in! It looked pretty good, especially considering how easy it was to get full-metal everything and the no work needed to "adapt" stuff to work like with the Savage suspension systems on a metal plate. It also cost less! The articulation was way more than anyhting I'd ever seen on a wheeled vehicle before which was good and the although there was zero-radius turning (left and right wheel directions cannot be reversed due to differentials/transmissions and other things like that), there was dual steering. Essentially there's no difference between front and back so zero-radius turning to get out of a jam is not really needed (you can always get out of a place the same way you came in right?!).

As for the motor controllers, I guess it's just like you guys. ONce you know too much about something, you can see all the drawbacks, everywhere that it could be easily improved, and all the unused potential. You know exactly what you want and more importantly, you know how to get it. It's the same for me and motor drivers. RC ESCs work great. But it's a problem when a computer is at the wheel since it's control to the electronics is not limited by a few switches and knobs. But it has no eyes for all that feedback for speed, overheating, stall, and other things like that. It needs direct access to all the speed/acceleration data from the ESC. It can also make use of the hundred different settings in the ESC all on-the-fly. You can also stick current and temperature sensors in the thing so the computer can know exactly how hard to push the ESC and not destroy it. It's also kind of a pain to sacrifice one line for every ESC and to also need to PWM it all the time. It's a lot more effective if you can just use a single serial connection to hook many ESCs to one line so you can set-and-forget speed, as well as get other data like speed, current, and temperature from the ESC. Not only that, you can send a hundred different commands to the ESC to use the hundred different settings all at once- not something you can do with a transmitter and your fingers.

The same thing goes for servos too. YOu can have many servos on a robot, and each one takes up one line and needs PWM constantly to function. With a custom PCB inside the servo, you can add a serial connection so you can control 40 servos from the same line, each with set-and-forget position control, not to mention adding sensors so you can also receive current/voltage/temperature feedback to push the servo to the limits without destroying it. The computer can also use that same line for servo speed and acceleration control on-the-fly as well as motion-control algorithms.


I kind of got a few projects that hve to get done at the same time and I'm already busy so I'm not sure how long it will take to get to everything- the brushless motor controller, custom servo PCB, a radar-like sonar thing to navigate for the robot, a main board for the robot (by far the easiest!), maybe a inertial/balancing sensor boards for the robot, and of course, I have to finish this piece of junk "motors-bolted-to-a-plate" robot I have right here that's super overbudget in time and money, with unimpressive results. That makes me mad. Not sure whether I should do the sonar, brushless ESC, or servo PCB first. THe servo PCB works fine in most cases so that's probably going to be the last of three. The custom ESC could be replaced by RC ESCs for a short-time, but as soon as it is replaced, it cannot be used for the robot anymore (unlike the servo where all the servo mechanics are still used).

So yeah, I had to type all that...TWICE.

[Reflection]

There was a guy that posted a build on here a good while back. It was a tank type thing or something. He was trying to build a 6 or 8 wheeled crawler thingymabob

It was running 3 or 4 E/T maxx front ends I think with some sort of belt drive system in the mix. I can't remember for the life of me who it was though. Might search for it or maybe someone will see this and know where it's at.

[DKNguyen]

Yeah, I ran by that and a couple others. Two with belt drive and one with shaft drive. None with independent steering on the wheels. They were all the E-Maxx suspension bolted to a plate.

Then there was that one rock crawler with the third pair of jurry rigged wheels on the back lol. I'd like to do that too, but it'd just agravate the steering. What I need is a longer wider chassis for mounting more electronics.

[WheelChair]

So, let me get this straight.

You are trying to backwards engineer a mamba max and then have it installed on a "universal" type PC board?

To save space?

[JohnRobHolmes]

He wants a sensored outrunner, and it needs to be compatable with autonomous software control. The mamba max wont do that.

[P47TBolt]

Originally Posted by DKNguyen

Crap...it logged me out while I was typing so I have to retype all this...

Yeah, university. I'd like to think my money got me something that Ican use for enjoyment. So here's my rant:

Since I started with robots, I've wanted to build a robot that's not limited by where it can go (like everywhere a person can). I use stairs as my benchmark since those are some of the hardest things to climb up and it really limits where a robot can go since stairs are everywhere,

Of course, robots with just two drive wheels that can only travel on flat ground (Segway-type two-wheeled balancers being the exception) were off the table right away since they can't go anywhere except for artificial environments. Segways were neat, but they lack expandibility (you can't exactly stick anyting new you build onto it because of the way it works and the mechanics of the design). They can't handle really rough terrain either. They travel fast though!

I started with Mars rover-type robots in mind. Those robots are totally impractical for hobbiests since they have itty bitty super expensive, heavily geared motors with advanced harmonic gearboxes that are shoved into a giant hollow custom-machined wheel that can contain the entire motor. As if that weren't enough, the motor also has to be able to support all the weight of robot since that's just the way it works. *Deep breath* and not only that each wheel is mounted to it's own little servo steering mechanism that also supports the weight of the whole thing!

Snakes, and legged robots were not very practical either. They can go many places but are slow, complicated, expensive, lack expandibility (it's kind of hard to build new experimental parts to go onto a snake since there are so many mechanical limitations). They are also power hungry, delicate, and heavy.

Tracks were neat. But smaller than life-size, larger than toy-size tracks don't really exist. The few that do are all made of plastic and rubber. They're expensive hard to work with, have a ton of parts that usually have to be customed and yeah...

With what we have, wheels are almost the only way to get good speed, efficiency, and load carrying capability. Most ATR robots you see in hobby are not much more than motors bolted to a metal plate. The lack of suspension and clearance really limits things. So the only thing left was all-terrain wheel vehicles that weren't as complicated as the Mars Rover suspensions, but had to be a lot more robust than merely some motors bolted to a metal plate. Of course, I'm talking about suspension but that's pretty complicated stuff that can get expensive- except in RC. RC is about the only place you can find mechanical stuff like that at good prices.

So for the longest time I was planning to rip the suspension off of HPI Savages and bolt 6 of them to a metal plate to get a 6x6 all-terrain rover. Each wheel would have it's own steering, and direction control to get skid-free zero-radius turning. But I wanted all metal parts too and we all know those cost a fortune especially if you need to get every part separately. It would have been expensive and a lot of work to adapt that all.

So I wandered the net and came across "rock crawlers". I didn't even know such a thing existed. Later I found out there were actually full-sized versions that people actually rode in! It looked pretty good, especially considering how easy it was to get full-metal everything and the no work needed to "adapt" stuff to work like with the Savage suspension systems on a metal plate. It also cost less! The articulation was way more than anyhting I'd ever seen on a wheeled vehicle before which was good and the although there was zero-radius turning (left and right wheel directions cannot be reversed due to differentials/transmissions and other things like that), there was dual steering. Essentially there's no difference between front and back so zero-radius turning to get out of a jam is not really needed (you can always get out of a place the same way you came in right?!).

As for the motor controllers, I guess it's just like you guys. ONce you know too much about something, you can see all the drawbacks, everywhere that it could be easily improved, and all the unused potential. You know exactly what you want and more importantly, you know how to get it. It's the same for me and motor drivers. RC ESCs work great. But it's a problem when a computer is at the wheel since it's control to the electronics is not limited by a few switches and knobs. But it has no eyes for all that feedback for speed, overheating, stall, and other things like that. It needs direct access to all the speed/acceleration data from the ESC. It can also make use of the hundred different settings in the ESC all on-the-fly. You can also stick current and temperature sensors in the thing so the computer can know exactly how hard to push the ESC and not destroy it. It's also kind of a pain to sacrifice one line for every ESC and to also need to PWM it all the time. It's a lot more effective if you can just use a single serial connection to hook many ESCs to one line so you can set-and-forget speed, as well as get other data like speed, current, and temperature from the ESC. Not only that, you can send a hundred different commands to the ESC to use the hundred different settings all at once- not something you can do with a transmitter and your fingers.

The same thing goes for servos too. YOu can have many servos on a robot, and each one takes up one line and needs PWM constantly to function. With a custom PCB inside the servo, you can add a serial connection so you can control 40 servos from the same line, each with set-and-forget position control, not to mention adding sensors so you can also receive current/voltage/temperature feedback to push the servo to the limits without destroying it. The computer can also use that same line for servo speed and acceleration control on-the-fly as well as motion-control algorithms.


I kind of got a few projects that hve to get done at the same time and I'm already busy so I'm not sure how long it will take to get to everything- the brushless motor controller, custom servo PCB, a radar-like sonar thing to navigate for the robot, a main board for the robot (by far the easiest!), maybe a inertial/balancing sensor boards for the robot, and of course, I have to finish this piece of junk "motors-bolted-to-a-plate" robot I have right here that's super overbudget in time and money, with unimpressive results. That makes me mad. Not sure whether I should do the sonar, brushless ESC, or servo PCB first. THe servo PCB works fine in most cases so that's probably going to be the last of three. The custom ESC could be replaced by RC ESCs for a short-time, but as soon as it is replaced, it cannot be used for the robot anymore (unlike the servo where all the servo mechanics are still used).

So yeah, I had to type all that...TWICE.

:? HUH? Just kidding.
Talk to JIA(JasoninAugusta).He is an electronics guru.If
anyone can help you,it would be him.

[DKNguyen]

Originally Posted by WheelChair

So, let me get this straight.

You are trying to backwards engineer a mamba max and then have it installed on a "universal" type PC board?

To save space?

Haha, no. Not to save space over the Mamba Max, but to save space over itself if done normally. This would be on custom PCB though. I'm asking RC4WD right now if they can custom make some longer chassis plates. That'd be quite a bit better than building the ESC smaller (since I could build everything else bigger too).