Renard based Servo controller [Archive] - Do It Yourself Christmas Forums

ctmal

06-11-2010, 09:14 PM

Here's something I've been working on. It's a Renard based servo controller. I ran the initial(very limited) tests on it tonight and it's working! I do want to redo it a bit to make it easier to etch and to make the headers more secure. I also need to test it in line with a Renard controller to make sure that works but I won't get to that until next week.
I tried to create the controller using similar parts to the Renards. It uses a pic16f688, the same oscillator, rs485 chips, etc. as the Renard SS. Right now it's running at 57600 baud but I've only tested 2 servos at a time(that's all I have right now). As long as the rest of the testing goes well it will be able to control 8 servos.
I also need to clean up the code a bit. The UART section I've just taken from the current Renard code and modified it bit(thanks for doing the hard part Phil:)).
Anyway, enough rambling for now.

5/27/2011 - I have placed an updated version of the Renard Firware in the File Library. If you're using the firmware in this controller there is no need to update the firmware but if you're using the firmware in a Renard controller the update might be beneficial for you. The new software includes remapping the outputs to the origional Renard outputs and allows for Negative outputs.

bobkeyes

06-11-2010, 09:36 PM

That's great! Just what I need. Will you make it available to the forum? I'd like to make about 6 of them.

Thanks and great job.

scorpia

06-11-2010, 09:41 PM

Thats an awsome little board,

i will be interested in seeing more information once you have tested it some more.

Peter

ctmal

06-11-2010, 09:44 PM

That's great! Just what I need. Will you make it available to the forum? I'd like to make about 6 of them.

Thanks and great job.

Absolutely! I just need to clean it up and test it some more.

ctmal

06-11-2010, 09:44 PM

Thats an awsome little board,

i will be interested in seeing more information once you have tested it some more.

Peter

Thanks, I'll put all everything up once I clean it up.

scorpia

06-11-2010, 09:54 PM

Thanks, I'll put all everything up once I clean it up.

just for an FYI,

If you can make the board under 5cmx5cm (2inx2in) then seeedstudio pcb service is a cheap way of getting a small run of boards made up for testing.

Peter

lbro

06-13-2010, 12:48 AM

How are you planning on using th board? Do you have something you needed it for?

ctmal

06-13-2010, 01:33 PM

How are you planning on using th board? Do you have something you needed it for?
My neighbor has challenged me saying she could out decorate me for Halloween. As long as I have time I'm planning on making either a talking skeleton or some kind of talking monster.

ctmal

06-13-2010, 07:47 PM

I was able to get some testing done. The results were great. I had the Servo controller connect to the computer then an XMUS Ren 16 connected in line. All worked as you would expect. The +12 volts and Ground were passed to the XMUS board with no issue.

I also cleaned up the code a bit and made the pads on all the resistors a bit larger to make it easier to solder. I'll have to etch another board to be sure but those changes shouldn't effect anything.

I'm going to post everything(in a couple of additional posts) if anybody is interest in playing/testing.

As always, and comments, suggestions or criticisms are welcome.

Chris

ctmal

06-13-2010, 07:53 PM

Attached is the code and the BOM I used. You see that almost all of the parts are the same numbers as the Renard SS series Boards.

A note about the bom. For the header for the Servos I just used header I removed from an old motherboard. I was looking on Mouser and I think 538-10-89-7061 would fit the bill but I'm not totally sure so if anyone can chime in on that it would be appreciated.

Basically what I've done is take the origional Renard code and modified it for use as a servo. I did not write the Rx routines, etc. myself, that was all from the origional Renard code.

ctmal

06-13-2010, 07:59 PM

budude

06-13-2010, 08:13 PM

Here's the board. Note there are two jumper wires on there. I wasn't really sure how I wanted to notate them on the silkscreen so I didn't do it;)

The DC input is 5V and the board will pass on the Zero cross signal if it's on port 3 and also the +12DC and Ground if present as in the xmus board setups.

Any question, give a shout.

Does the PIC source/sink enough current to drive the servos? I don't know much about servos but was wondering if a driver (ULN2803, transistor, etc) is required? I think the PIC can sink 90mA max for all outputs (something like 15mA per single output with derating?).

ctmal

06-13-2010, 08:54 PM

I wasn't able to find any documentation but I measured the current in the signal wire to the servo. The highest I came up with way .036 ma so it shouldn't be a problem.

ppohlman

06-14-2010, 10:48 AM

hey ctmal -

This is great!! Thanks for putting this together.
omzig had started something similar in this thread (http://www.doityourselfchristmas.com/forums/showthread.php?t=6597), but he got busy and had to postpone his project. I will definitely be looking at doing some 3-axis skulls as well.

ctmal

06-14-2010, 11:08 AM

Nice, I can't wait to hear how it works out.

jmksparks

06-14-2010, 11:25 AM

Is it possible that this would interface with a "linistepper". I am in the process of biulding this servo/stepper Driver for 1amp motors and it would be nice to control with Vixen .

P. Short

06-14-2010, 01:21 PM

The control signal for hobby servos is just that - a control signal. The current requirement is quite low, and I'm quite surprised that it is as high as the 36mA that ctmal measured.

ctmal

06-14-2010, 01:39 PM

The control signal for hobby servos is just that - a control signal. The current requirement is quite low, and I'm quite surprised that it is as high as the 36mA that ctmal measured.

It was .036ma that I read. Not totally sure of the accuracy since the batteries in my multimeter need to be replaced but well within the limits of the pic.

Is it possible that this would interface with a "linistepper". I am in the process of biulding this servo/stepper Driver for 1amp motors and it would be nice to control with Vixen .

I'm not familiar with the "linistepper" so I can't really comment on it. This board just supplies the +5v, Ground and signal to control the servos.

jmksparks

06-14-2010, 03:44 PM

I'm not familiar with the "linistepper" so I can't really comment on it. This board just supplies the +5v, Ground and signal to control the servos.[/QUOTE]

"www.romanblack.com" version 2 of his linistepper for test purposes they are using 555s. If you are able to purouse this unit and give me an opinion would be much appreciated. John.

ctmal

06-16-2010, 06:28 PM

I'm not familiar with the "linistepper" so I can't really comment on it. This board just supplies the +5v, Ground and signal to control the servos.

"www.romanblack.com" version 2 of his linistepper for test purposes they are using 555s. If you are able to purouse this unit and give me an opinion would be much appreciated. John.[/QUOTE]

Again, I'm not really familiar with it but it doesn't look to me like would work. That board appears to accept CNC protocol input from a computer, not server signals.

jmksparks

06-16-2010, 11:36 PM

"www.romanblack.com" version 2 of his linistepper for test purposes they are using 555s. If you are able to purouse this unit and give me an opinion would be much appreciated. John.

Again, I'm not really familiar with it but it doesn't look to me like would work. That board appears to accept CNC protocol input from a computer, not server signals.[/QUOTE]

Hope its not to obviouse that I do not know what I am talking about, thank you for looking. I thought it was a " driver / booster " maybe I should ask the propeller guys if this would work as a booster for higher amp motors on their unit ?.

ctmal

06-23-2010, 09:58 PM

I just got back from my mom's house where I had my old radio controlled plane. I stripped it of it's futaba servos to give them a test. Good News! They worked great. Actually they seemed to work even better than the Hitec HS-55's that I was using before.
Still waiting for my skull to come in to actually make something usefull with this but so far both Hitec and Futaba servos are working.

ctmal

10-20-2010, 11:47 AM

Just an update...
I'm running way behind on getting sequencing done for Halloween(and anything done for Christmas!) but I do have the talking skull hooked up to the servo controller and all seems to be working great. So far I have the jaw movements programmed for a song, I just have to work in the head movements. Then I get to put it out.
When I get it all sequenced I'll have a friend video it for me and post it.

ErnieHorning

10-20-2010, 01:31 PM

This is pretty cool Chris! I kind of remember seeing this thread but didn’t think much more of it because I wouldn’t be here for Halloween. Since you only modified the output of the Renard code, this could replace the code in anyone or all of the Renard 64 micros. It would be pretty easy to replace a remote SSR.

I ran across a servo driver board that could drive a wiper motor. It used a home made optical decoder for the shaft position feedback. I can’t find what I did with the instructions though I probably still have it somewhere.

I can see these being useful for both Halloween and Christmas.

OK I think I’m already planning for next year and I haven’t started putting up this years display.:D

P. Short

10-20-2010, 04:25 PM

Now to get more vixen support for it...it would be nice if the preview could show the object that is being controlled and the movement of that object.

ctmal

10-20-2010, 04:39 PM

Now to get more vixen support for it...it would be nice if the preview could show the object that is being controlled and the movement of that object.

That would be a huge help. As it stands now I need to have it plugged it in order to sequence.

Another huge help would be a joystick input to sequence the lights. Even better would be something that has 2 joysticks like a radio control airplane transmitter.

ctmal

05-27-2011, 05:13 PM

An update to the firmware has been posted in the File Library. There are some details on the update in the first post.

gdyrdave

11-04-2011, 08:38 AM

Can this firmware be modified for having a start address?

Thanks
Dave

dirknerkle

11-04-2011, 09:39 AM

Can this firmware be modified for having a start address?

Thanks
Dave

I don't know if it has that capability, but a workaround could be to use an SS8 controller WITH a start address and daisy-chain that into the servo controller. Out of a 16-channel stream, the SS8 would take the first 8 and pass the remaining 8 on to the servo controller. You'd probably want to use some lights anyway, so the 8 channels of the SS8 wouldn't be wasted, unless of course, you just chose to use it for the start address only and not use its 8 channels...

LabRat

11-04-2011, 09:48 AM

Would a DMX 8 channel servo firmware for the 16F688 be of any use?

ctmal

11-04-2011, 10:23 AM

I don't know if it has that capability, but a workaround could be to use an SS8 controller WITH a start address and daisy-chain that into the servo controller. Out of a 16-channel stream, the SS8 would take the first 8 and pass the remaining 8 on to the servo controller. You'd probably want to use some lights anyway, so the 8 channels of the SS8 wouldn't be wasted, unless of course, you just chose to use it for the start address only and not use its 8 channels...

The firmware by default runs like the native Renards without a start Address.

Nice workaround though.

gdyrdave

11-04-2011, 10:53 PM

I put the firmware in the last 688 in a ren48lsd. Works like a champ. Now to make a servo interface card. Wish the addressing aspect of the firmware could handle this. used an ren8x8xb for testing before i put it in the 48lsd. Want to modify a 8x8xb just for servos. Thanks guys for the ideas.

Dave

soakes94

11-07-2011, 03:36 AM

Would a DMX 8 channel servo firmware for the 16F688 be of any use?

Did you ever write the DMX code? SInce I would like to program one of my ren48lsd chips with the code but I'm running my ren48lsd on DMX.

Steve

LabRat

11-07-2011, 06:51 AM

Yes, DMX in, and sends a series of pulses (1-2ms) based on the channel value (0x00 to 0xFF)

I've been trying to find time to write a DMX Address programming option into the FW, but it's not ready for primetime
just yet, so you would have to write the address into the EEPROM manually. This is trivial if you have a PICKIT, and
a little more hassle if you don't.

I'm optimistic that I will have two "robotic" lights running under my eaves this year.

soakes94

11-07-2011, 01:24 PM

Could you send me a copy of the code please? And I will give it a go when I have some spare time?

Thanks Steve

soakes94

11-08-2011, 09:17 AM

Ive programmed my pic, put in a ren48lsd and cant get anywhere, when i connect my servo i get a small amount of movement but thats it, any ideas?

LabRat

11-08-2011, 12:21 PM

Did you update the first two bytes of the EEPROM memory space to be the start address (stored as HEX)

eg. Start Address 100 = 0x00 0x64
Start Address 256= 0x01 0x00

ctmal

11-08-2011, 03:39 PM

Ive programmed my pic, put in a ren48lsd and cant get anywhere, when i connect my servo i get a small amount of movement but thats it, any ideas?

Is this using the Renard Servo firmware?

LabRat

11-08-2011, 03:50 PM

Issue was resolved.. the start address wasn't programmed into the EEPROM.

This is using DMX based Servo control FW that I created. Seemed to be working, but all channels seem stuck on 4.58v now. (after turning on the "laser ??")

(Not my laser... just reporting what I was told)

I'm suspicious that the transistors may have been blown.

soakes94

11-08-2011, 04:55 PM

My ren48lsd is now dead :( All my pics are outputting 5V, doesnt matter if the channel is set at high or low, anyone got any ideas? Im going to start trying to reflash everything in the hope that doing that will fix it

soakes94

11-08-2011, 05:28 PM

Ive reprogrammed all the pics and nothing, I tried a working pic from another board and nothing, but that pic works in the other board, could it be my crystal? Is there a way to check? I dont have an oscilliscope, just a multimeter haha

budude

11-08-2011, 06:00 PM

Follow the troubleshooting tips in the Ren48LSD wiki and see if anything hits there!

gdyrdave

02-03-2012, 06:57 AM

Chris,

Is there a way to set limits in the firmware. I know the pulses are .5 to 2.5 ms but can we set the span to like .75 to 2.3 or something? my servos max out and I broke a gear in one already. I set the limit in Vixen but somehow when it goes to 0 it drives the servo to one end and chatters. I would be nice to hardcode this in firmware.
Thanks

Dave

ctmal

02-03-2012, 07:54 AM

kychristmas

02-03-2012, 07:15 PM

Currently there is no way to do that in the firmware.

If you set a cell to 0 then you've gone past any limit's you've set. In other words, if you sent a 0, the firmware pulses a .5. If you need a pulse higher than that you need to send a value higher than 0.

If you were using 2.5, could you use dimming curves to accomplish the limits? Where 0 to 30 would result in .75 and maybe 90 to 100 is 2.3. Not the right numbers, but hopefully you get where I'm going.

ctmal

02-04-2012, 11:33 AM

If you were using 2.5, could you use dimming curves to accomplish the limits? Where 0 to 30 would result in .75 and maybe 90 to 100 is 2.3. Not the right numbers, but hopefully you get where I'm going.

I dont see why that wouldn't work. I've never played around with dimming curves though.