Open-Encoder-MG995

Discussions relating to development and use of the OpenServo hardware.

Moderators: jharvey, Secondary Admin, Admins

BasicFox
Posts: 59
Joined: Sun Mar 15, 2009 2:45 pm
Location: Belgium

Post by BasicFox » Tue Oct 13, 2009 10:43 pm

if you use the headers I2C each openencoder has to have an other adress and then you would have 2 masters on the I2C bus (OS controlling the OE and the OSIF controlling the OS)

haven't looked over the code but with a routine thats has the clock and data pin as parameter would ideal

ginge
Site Admin
Posts: 1031
Joined: Sat Jan 14, 2006 2:34 pm
Location: Manchester, UK
Contact:

Post by ginge » Tue Oct 13, 2009 10:47 pm

Well, the OpenServo doesn't support multimaster, and without going into detail, multimaster on AVR is flakey at best.

Also having each OE and OS on one bus would make it noisy and crowded, especially for the sample rate we need (10ms) for the PID.

jharvey
co-admin
Posts: 362
Joined: Sun Mar 15, 2009 12:06 pm
Location: Maine USA
Contact:

Post by jharvey » Wed Oct 14, 2009 8:43 am

Right now my plan is to bit bang it on it's own sub buss, using SDA as PC2, and SCL as PD5. Here's a picture.

Image

I'm planning to bit bang with the current code, not the -dev code.

robotjay
co-admin
Posts: 225
Joined: Tue Aug 01, 2006 3:58 am
Location: Nebraska, USA
Contact:

Post by robotjay » Mon Nov 23, 2009 12:02 pm

Jared,

Did you ever buy custom shafts like you mentioned in another thread?
jharvey wrote:On a down side, pricing for the bronze shaft replacement is stupid expensive. So far the best price has been emachineshop.com. In qty 100, they charge about $4.50 each, in qty 50 they charge about $7 each, in qty 10 they charge about $30 each... Because I should be able to glue the first ones in, I'll hold off a bit before I order the replacement shafts.
Have you considered using rapid-prototyping to create the prototypical shafts? If you arrange the shafts on a tree-shape (as pictured below) you can pack a ton of parts into one cheap piece. (Yes, I took this idea from Cliff Huston, who designed the OSv3.) I took your POT_shaft_B.sldprt file, arranged it into a grid, and had it quoted by quickparts.com. If the parts are made of standard SLA material, for $150 we can have 114 custom shafts by next week. For $188 we can have them tomorrow. (Also, the 19x6 grid size I chose was arbitrary. It is likely we can increase the grid size a good deal, without affecting the price.)

Image

Also, could we spec a press-fit for the magnet end of the shaft, instead of the "rivet" hole? Seems easier to locate the washer that way, as opposed to using glue or mechanical fasteners. Or is the magnet ID not closely toleranced? Talk to you soon.

-Jay
"Nothing is fool-proof; For we fools are ingenious and will find a way."

jharvey
co-admin
Posts: 362
Joined: Sun Mar 15, 2009 12:06 pm
Location: Maine USA
Contact:

Post by jharvey » Mon Nov 23, 2009 9:58 pm

So why did you have to do that after I assembed a handfull of them :lol:

I have not purchased the custom shafts. Also I hadn't thought of the plastic idea. I'm not in a rush, so I don't mind keeping the costs down. Is this something you're thinking of doing up? I could pitch in say $50 ish.

I like the interference fit idea, perhaps we could make it a spline like shaft, and let it really mash together.

Sounds like you have access to Solidworks?

robotjay
co-admin
Posts: 225
Joined: Tue Aug 01, 2006 3:58 am
Location: Nebraska, USA
Contact:

Post by robotjay » Tue Nov 24, 2009 11:56 am

So why did you have to do that after I assembed a handfull of them
My deepest apologies. :) If I had seen this thread sooner, I would have come running.

I like the interference fit idea, perhaps we could make it a spline like shaft, and let it really mash together.
It seems like you have a lot more formal schooling than me. Do you mind determining the right dimension to give us a good snug fit? We could use a spline if you really think we need that much holding power, but I don't think it's even necessary. The only load on the shaft will be the rotational inertia from the washer itself, and gravity.
Sounds like you have access to Solidworks?
Of course. :) As a robot geek, I'd lose all my street cred without it.
I could pitch in say $50 ish.
I'll match that. If we get a third person in on it, (Barry. FireBALL. I'm looking in your general directions... ;)) It'll be super cheap to have a ton of these lying about. (Imagine: no more hand-drill, C-clamp, ghetto-rigged lathing necessary. ;))

Talk to you soon.

-Jay

EDIT 11/24/2009: Not that I'm not impressed by your ingenuity in using a drill in this way. (+1 rep for that idea.)
"Nothing is fool-proof; For we fools are ingenious and will find a way."

jharvey
co-admin
Posts: 362
Joined: Sun Mar 15, 2009 12:06 pm
Location: Maine USA
Contact:

Post by jharvey » Tue Nov 24, 2009 5:42 pm

I have some learning to do, but it looks like I can expect +/- .005 inches. I'll see if I can make a draft shaft this weekend / Turkey day.

jharvey
co-admin
Posts: 362
Joined: Sun Mar 15, 2009 12:06 pm
Location: Maine USA
Contact:

Post by jharvey » Thu Nov 26, 2009 8:43 am

It was brought to my attention that I didn't release the drawings for the magnet. I posted it, as well as the Solidworks files for the OE pot. They are linked to from this page.

http://www.openservo.com/ConstructionTu ... en-encoder

See the "Reassemble POT with the OE board" section. Enjoy.

FireBALL
Posts: 28
Joined: Fri Jul 24, 2009 4:31 am

Post by FireBALL » Sat Nov 28, 2009 5:20 pm

ok here it is
Barry. FireBALL. I'm looking in your general directions...
I'm already in a dialog with a company for procuring pot shells and the spindles, I also have a desktop lathe machine of my own but I'll be more than willing to contribute USD$25 towards the cause or I'll share US$50 if Barry sends me a OS3v board.

What do you think Barry? is it fair??

cheers

jharvey
co-admin
Posts: 362
Joined: Sun Mar 15, 2009 12:06 pm
Location: Maine USA
Contact:

Post by jharvey » Sat Nov 28, 2009 5:55 pm

It appears that one of my OE's is failing. Seems to drift a bit. If I rotate it say 90CW and back, I don't get the same number anymore. This likely indicates the glue has given way.

So my interests in this shaft idea are growing. Are we still looking at http://quickparts.com/, or do we perhaps have another vendor that might be beneficial?

jharvey
co-admin
Posts: 362
Joined: Sun Mar 15, 2009 12:06 pm
Location: Maine USA
Contact:

Post by jharvey » Mon Nov 30, 2009 11:22 am

Robotjay, can I get you to call quickparts.com during business hours? I'm not able to do it. There are some things I'm curious about that I think they can answers. Also I'll put up around $100 if need be.

I wonder what are the mechanical properties of the SLA materials they use. Young's modulus ect.

You drew a runner, I don't think that's required for SLA. SLA often stands for http://en.wikipedia.org/wiki/Stereolithography which isn't injected, but printed. Perhaps we can lower the cost by simply making an array of parts that are floating in space. Standing them on end would likely reduce the need for support material.

They offer several methods, perhaps SLA isn't the one they might recommend. Their page notes a call can often ensure the process will be the right process for the application.

I see a tolerance of +/-.005 inches. Does that hold true for such small parts, or could we expect a tighter tolerance because of the small size?

robotjay
co-admin
Posts: 225
Joined: Tue Aug 01, 2006 3:58 am
Location: Nebraska, USA
Contact:

Post by robotjay » Wed Dec 02, 2009 10:13 pm

jharvey,

I asked QP if they'll allow us to send them a single STL file with multiple, free-floating parts, and the sales rep said he'd allow it just this once. So I arranged them into a 15x15 grid, stood the parts on end, and the price came back as follows:

$188.00 - normal resolution ABS-like material, 3-5 day lead time.
$395.00 - high-resolution ABS-like material, 3-5 day lead time.

The physical properties of the materials they use can be found on this table: SLA Material Properties.

The sales guy recommended ABS-Like White (Somos 18420) for our application, but if you like a softer material (or a "stretchier" one) it's your call.

He didn't answer about the .005" tolerance, so I take that to mean the tolerance will be the same no matter the size of part. Based on my description of what we're doing with these shafts, he still recommended the SLA method. I mentioned that we'll be pressing these into washers, and he said that if we would rather use a softer material, that cast urethane would do nicely, but it would increase the price ten-fold.

ALSO, the sales rep said that the parts would come back "rough." Meaning, that there will be small, grain-of-sand sized bumps on the surface that would normally be cleaned off by their technicians. Because of the way we're cheating the system to get so many parts at once, they won't bother cleaning them. If we want a smooth finish out-of-the-box, we'll have to pay per individual piece, which dramatically affects the price. I imagine that going over each piece with the polishing tool on my Dremmel won't be a problem.

Let me know where you want to go with this, and I'll get the parts ordered. Talk to you soon.

-Jay

EDIT 12/02/2009: The bumps I'm talking about can be seen in this photo:
Image
"Nothing is fool-proof; For we fools are ingenious and will find a way."

jharvey
co-admin
Posts: 362
Joined: Sun Mar 15, 2009 12:06 pm
Location: Maine USA
Contact:

Post by jharvey » Thu Dec 03, 2009 9:33 am

About those bumps, that's not a huge deal. Some times you get some that are held on very lightly, and will come off by simply rubbing your hand over it. Others may require some work with a razor blade to pluck them off. I agree we want to do that if it helps keep the cost down. In our case, with it on end, I think we would only get one, that we probably don't care about. I would recommend against a dremel, even with a polishing wheel, the high RPM creates a lot of heat, and will very likely damage the part. We can of course try it, I suspect a razor blade will work much better.

I'll look over the material specs they listed on that page. I'll let you know if I think we can get parts to work with the .005 and those hardness specs.

They didn't spec what high and standard resolutions is. I wonder if .005 is high resolution, or if they can hold a tighter tolerance.

Also I think we may want to have some custom tools made as part of this process. We may want a fixture of sorts that we can use to hold the shaft, pot case, and magnet when pressing these together.

He didn't happen to mention how their pricing structure works did he? I wonder if they charge say $150 for setup, then $38 for the plastic. If we half the batch size, could we potentially save significant bucks, or is it just a buck or two?

BasicFox
Posts: 59
Joined: Sun Mar 15, 2009 2:45 pm
Location: Belgium

Post by BasicFox » Thu Dec 03, 2009 1:50 pm

I have my first readings out of the OSv3, finally :p, after all the trouble trying to solder the µC onto the PCB.

I'm very interested in the PLA method, this way the magnet will be mounted much better on the shaft. And since there is no friction the part doesn't have to be that strong, count me in :p

are there allready pin's designated for the OE? so we all start off with the same setup.

Now I can do all the programming on one OS with jahred his PCB's. But a special pcb grid where there is allready a OE pcb included and and easy 4 wire connection holes would be ideal for my other 20 servo's that need to be converted into OS :D (future plans)

robotjay
co-admin
Posts: 225
Joined: Tue Aug 01, 2006 3:58 am
Location: Nebraska, USA
Contact:

Post by robotjay » Sat Dec 05, 2009 12:35 pm

jharvey,
jharvey wrote:He didn't happen to mention how their pricing structure works did he?
No. He didn't describe their pricing system. What I've observed from their online quoting system is that regardless of the part size, you're gonna pay $150 minimum. It's likely you're right in that there's a $150 setup fee, plus material. The cost for just one shaft is $150. The cost for 114 is $150. The cost for 225 shafts is $188. Go figure.

As you've mentioned before, it's entirely likely that we'd save some money by choosing a different shop. I chose quickparts because they have an easy, convenient, and "instant gratification" quoting system right on their website. You're welcome to shop around.

As far as creating extra tooling, I would hope that we could simply push the magnet onto the shaft without any tools, but maybe .005" tolerance is too loose to be reliable without some sort of press and centering device. Expect to pay more to add things like that to the final part file.

-Jay
"Nothing is fool-proof; For we fools are ingenious and will find a way."

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests