31 replies to this topic

[Les Newell]

Quote

That's why I recommended SuperModified servos.

I just looked at that link. They fit an encoder which is pretty much what I said. You don't really need their electronics because you can do the same thing with an Arduino and that motor shield.

[D-Lite]

Reddog, on 23 January 2012 - 12:47 AM, said:

An alternative is possibly the OpenServo group, which is open source. Has anyone played with them?

The SuperModified stuff is open source also and very similiar to OpenServo. OpenServo seems to be the more active project.  On the other hand, SuperModified is complete and working while some of the features like magnetic encoder feedback still seem to be work in progress on OpenServo. This is just my first impression when looking at both projects, haven't used any of them myself.

[D-Lite]

Les Newell, on 23 January 2012 - 07:31 AM, said:

I just looked at that link. They fit an encoder which is pretty much what I said.

Yes, you are right of course. I just wanted to express that this already exists as a complete project that fits into a standard servo and has all the features needed. In the end, their are many ways to address this problem and in the end it's a matter of taste and personal experience what's the most appropriate solution.

Les Newell, on 23 January 2012 - 07:31 AM, said:

I just looked at that link. They fit an encoder which is pretty much what I said. You don't really need their electronics because you can do the same thing with an Arduino and that motor shield.

The advantage is that Arduino and motor shield are available while with SuperModified, you'd have to make you own PCB. For OpenServo, there are PCBs available(currently out of stock) but only for the main board, not the encoder.

[Reddog]

D-Lite, on 23 January 2012 - 10:50 AM, said:

The advantage is that Arduino and motor shield are available while with SuperModified, you'd have to make you own PCB.

All your feedback has been brilliant, thank you!

Maybe I am missing something but the motor shield does not seem to have position feedback, so if the servo or stepper get out of alignment I will never know.

[Reddog]

D-Lite, on 23 January 2012 - 10:36 AM, said:

The SuperModified stuff is open source also and very similiar to OpenServo. OpenServo seems to be the more active project.  On the other hand, SuperModified is complete and working while some of the features like magnetic encoder feedback still seem to be work in progress on OpenServo. This is just my first impression when looking at both projects, haven't used any of them myself.

I cannot seem to see where you can get the schematic of the SuperModfied servo, eagle files and BOM. I checked their Google page, maybe I missed it.

[D-Lite]

Reddog, on 23 January 2012 - 03:06 PM, said:

I cannot seem to see where you can get the schematic of the SuperModfied servo, eagle files and BOM. I checked their Google page, maybe I missed it.

The schematics are in the datasheet. The PCB files seem to come from another user and are hosted here. Be careful to check if the PCB and code supports PWM input if you want to use that. I2C and UART are also possible and especially UART would be nice for a direct connection to a PC but I havent looked into the code and schematics if this is all supported at the same time or if there are different versions or compile time options that enable the different features.

[Reddog]

D-Lite, on 23 January 2012 - 06:53 PM, said:

The schematics are in the datasheet. The PCB files seem to come from another user and are hosted here. Be careful to check if the PCB and code supports PWM input if you want to use that. I2C and UART are also possible and especially UART would be nice for a direct connection to a PC but I havent looked into the code and schematics if this is all supported at the same time or if there are different versions or compile time options that enable the different features.

I have spent ages looking at Open Servo and the SuperModified boards. A couple of things I have noticed:

1. While it seems the Open Servo guys are more active (they exist) the activity is sadly pretty sparse. Their boards are not available and they are not sure when more are being built. Development of version 4 started some time in 2010 and its not mentioned much again after that in their forums.

2. It seems the encoder board (and the two other attached boards) for the SuperModified servo look like it would be pretty hard to mount properly unless the space where the Pot was (in the servo) was exactly the same width and breadth of the SuperModified boards. Aligning the board and magnet perfectly would be pretty hard and I have seen some comments about this on the web.

3. While the SuperModified is the most complete, there is no support at all. Any issues would be hard to fix, I would probably need to ask here and its not really the place to ask.

Its a bit of a shame somewhere like sparkfun or somewhere else has not picked up the SuperModified and started building the boards or at least continued on with the development.

To fix the SuperModified encoder issue I was thinking it might be worthwhile to use the Open Encoder from Open Servo and connect it to the SuperModified control and power boards. This would allow for a more flexible approach to mounting the magnetic encoder.

[Les Newell]

Reddog, on 23 January 2012 - 12:22 PM, said:

Maybe I am missing something but the motor shield does not seem to have position feedback, so if the servo or stepper get out of alignment I will never know.

With a servo you would have to implement the encoder feedback yourself. You would need to use two input pins and decode the quadtrature signals. I would have thought there is already an Aruino library available to do this. WIth a stepper you only really need a 'home' switch that tells you when the motor is in one specific position. In a simple setup you would only use this home switch once when you initialize but if you want to be clever you can check every time you pass the switch to make sure the motor is where you think it should be. For a tracker you shouldn't have to worry about lost steps too much. The stepper shouldn't be working all that hard.

[Reddog]

Les Newell, on 25 January 2012 - 03:10 PM, said:

With a servo you would have to implement the encoder feedback yourself. You would need to use two input pins and decode the quadtrature signals. I would have thought there is already an Aruino library available to do this. WIth a stepper you only really need a 'home' switch that tells you when the motor is in one specific position. In a simple setup you would only use this home switch once when you initialize but if you want to be clever you can check every time you pass the switch to make sure the motor is where you think it should be. For a tracker you shouldn't have to worry about lost steps too much. The stepper shouldn't be working all that hard.

Thanks Les, really appreciate the feedback!

[Reddog]

Out of interest I have been checking out the SuperModified servo boards and schematic. Something is weird about the Eagle board files and I was wondering if someone could explain please? Some of the components are off the board and I am not sure why. I have posted a screenshot.

 Screen Shot 2012-02-06 at 10.26.17 PM.png   24.97K   16 downloads

[Sambas]

Looks like work in progress, board isn't completed. All yellow wires are "rubberbands" that are not yet routed.

[Reddog]

Thanks mate. Only learning this stuff, its extremely interesting. Thanks for the help on this one.