Tronic's AMC motor motion-controller with pwm/servo output

MD03 looks good. It is one of the choices I would go after if I decide to go 24V vs 12V. Another one I am looking at is:

http://www.dimensionengineering.com/Sabertooth2X25.htm

There definately seems to be a rise in cost the higher voltage you go. For now I am going to stay 12V but I feel that is only going to go so far in the way of speed and response. I want to make sure 12V works first before burning out auto motors with 24V...many seem to think the auto motors will take it though

keep up the good work on the I/O board :top:

regards
Kyle

Hi Thanos,

I'm trying to get the pcb etched by a friend of mine (home made). He says the edges from the component contacts are to small, they will be etched away.

Is it possible to make a pcb with bigger edges on the component contacts for us home etchers


Thanx

Hi Rovor,

No, its not possible to change the size of the compoments contact (I assume you referring to the holes outline), as they are generated automatically from the PCB CAD program for the compoment requirements. Thats why some compoments have thicher holes etc.

What method are you using to etch the PCB? Have a look on the way, I sometimes, use to make mine:

http://ptyxiouxos.net//greekbotics/user ... gGuide.htm


In fact in the guides photos you can see the results of etching an older version of my controller!!!


Regards, Thanos

When I returned home yesterday, a nice surprise waited me. I found outside my door two parcels. One with my AVR controler PCB and the other was the order with the extra parts from RobotikHardware.



I was very surprised too by the quality of the PCB (tin-plated) as well the fact that was already drilled too!! WOW!!! Thanks Ego!!!!

In the photos you can compare the quality of the ordered PCB vS the homemade that I have build some months ago. Ok the homemade PCB is a little rusty now but you can compare the detail level!!! Incredible!

I quickly placed some of the parts on it show you what is going to look! I hope to have it soldered and tested this weekend!



The second parcel had the second RV-VN2 controller I needed, as well a set of optical encoders (120 CPR) to experiment with.



I hope to have the new controller soon installed on my joyrider for some true motion tests!!!

Regards, Thanos

They are cheap.

If you need more power, then use other drivers.

Well that takes me out of the running with the AVR board.

Thanos confirmed his board is not compatible with either the 28V/25A simple-H controller which I use or the KT-5198 Dual Bidirectional DC Motor Speed Controller (50V/10A+) which I have ordered.

I had hoped this would not be the case but it looks like I will have to find another solution. Higher voltage/current H-bridge controllers compatible with his board are simply too expensive!

LOL I have 4 etched boards to hang on the wall! :

EvanF

I could give you a adres to send them to....

All the way on the other side of the Globe.

:razz: thanks for your kind offer RoVoR I'll keep it in mind, I've calmed down a little from my disappointment, I've waited for so long so it was a shocker to hear I've got more issues because of my earlier choice to run with 24V.

Since I've already ordered it I'll give the KT5198 a go and also look at hacking my own H-Bridge with the IRFZ44 (49A); and the IRF4905 (74A) mosfets.

I'm sorry that I made you so sad about it. But I'll try to help you in another way.

As I seen your KT-5198 Dual Bidirectional DC Motor Speed Controller also has RC servo input. I will add support for servo output in the next beta version so you can use your controller atleast this way... And it won't be much slower in responce from using direct PWM input!
I already did this with my HB-25 motor controllers that has RC-servo input...

I hope to have the RC-servo output on my AVR controller, ready by the end of the week. I'm just too busy these days preparing to take part in a TV show about robotics (in the Greek National network, in greek called ERT!!). They also asked me to present there my joyrider simulator!!!


Regards, Thanos

The RN-VN2 limits the max voltage not the VNH2SP30 it uses.

One with the proper designing knowledges could desing a new motor controller based on VNH2SP30 that can handle the max of 41volts or the 25amp. But it will be some sort of expensive to do something like that since much measures must be taken to prevent overheating and allow the VNH2SP30 to provide the 25amp safely.

Thats why the developers of RV-VN2 choosen to limit these features.

One thing that prevent me fron designing such motor controller based on VNH2SP30 is that its surface mount compoment!!!


Regards, Thanos

CRITICAL AVR controller PCB FIXES

I started the soldering of the parts and noticed two critical design errors I made.

First is the orientation of the ALPS button that used for reset. I made wrong placement and if placed as diplayed in the silk layer it will cause constant reset, meaning the AVR will never boot to the code... I show on the next photos how to cut two of its legs and bend the other two to fit in to the board. I will fix this in the next version of the PCB design.








Second issue is that I mistaken the power line for the potetiometers and its connected in the ground instead. In result the pots will not read values at all!! I fixed that by cutting the wrong line on the PCB and jump wiring to the proper power line as seen in the photos below.








I haven't much time to solder all the compoments (Resistors and capacitors) but I soldered most of them. I had a little trouble though to mount the pin headers for the pots (with black bottom) as the pins was a little bigger than the holes on the PCB. I managed to insert the with little force using a flat screwdriver. And mounting the molex pinheaders (white base) neede also a trick. I bended its legs that doesn't have assigned holes on the PCB as seen in the photo below.









And something weird... For while I believed that I have burned the Pitch motor on my joyrider after I used the frex profile for the LFS the includes mixing the Z axis (heave) into the motion... After an hour of fantastic driving on dirt (to feel the bumps) it just stopped working and it was locked full forward. I got out of the sim to check what happened just to find out that the pitch motor was extremely hot! The bank motor was just a barely warm in contrast... After and about an hour I returned to examine better the sim to find out that the pitch pot had a broken wire causing the feedback signal to be zero. It must have been dettached by the fast motion... Now I hope that the motor wasn't damaged because it was forced to the end of the travel for over 5 minutes that I was trying to find what happened!

And if you wonder, I just used the old BS2 setup with the HB-25 motor drivers!!! Can't wait to mount on it the new avr controller it will be awesome!!!!!

Again, I shooted all this cool driving on dirt (LFS) with my videocamera to see yourself how it was. I will upload it soon to see it.



Regards, Thanos

That's awesome news! Has anyone told you, you're a great guy!?

There's no need to apologise, it's not your fault, we're all experimenting and it's trial and error all the way, it always is in a relatively new field. I'm sure we've all pumped a lot of money into simulator stuff. 12V or 24V, well we now know one difference and it's $

I'd like to know how fast mosfets can switch, whether there is a limit to how high PWM frequency they respond until. I see a lot of intergrated h-bridges now and wonder why the trend (maybe smaller size/weight for robotics?). They're more expensive and if they pop then harder to replace than say a single mosfet in a bank of 4.

So you are a star on TV now hey! 8) .. don't forget us

EvanF

I uploaded the new video in which I use the frex-like profile for LFS that mixes the z-data into the motion.

http://www.youtube.com/watch?v=d1BpKMXlvOw

Watch it closely, it won't display alot of motion but I could feel the extra bumping for sure. The bump effect is more intense after the 3rd minute of the video that the suspensions of the car was damaged enouph to do not absorb much of the bumps!!

Regards, Thanos

Your sim seemes to work very well. :top:
How does it track the target value? Is it possible to record a little video of the target/sim graph in the profiler?

Well, my controller does externally the tracking of the target values. If I shot a video to see it you will always see the sim and the target values to be the same!

Its totally different from your pneumatics control method since my controllers does not return the position back to the x-sim software. They have their own way to calculate the position feedback.

So, the x-sim works for me in blind mode (when using USO). It only output the position values hoping the external hardware will handle the position correctly.

Thanos

If you can't find them, its ok, you can always use the normal pin headers with black base. But then you have to be carefull for the orientation of the connector.

Regards, Thanos

RaceRay,

Take a look here, they have a lot of molex connectors

http://de.farnell.com/jsp/search/browse ... stid=17484