Back
[02:26:08] <SpeedEvil> codepython777: hoverboard wheels
[02:26:30] <SpeedEvil> 600RPm or so tops. not 12V though
[09:48:09] <codepython7771> SpeedEvil: thanks
[16:25:36] <branjb> what's a good strategy with dealing with degrees/radians in a PID loop around zero? ie, going from 359 degrees to 0 degrees makes a sudden error jump that's not real
[16:26:41] <deshipu> use modulo arithmetic?
[16:28:04] <branjb> !
[16:28:10] <branjb> duh
[16:30:20] <z64555> does that cover negatives?
[16:30:34] <z64555> I haven't done modulo with negatives before
[16:30:57] <branjb> probably
[16:31:12] <branjb> i'll have to parse it anyway since my theta is a float
[16:31:48] <z64555> huh. google says it does
[16:31:54] <z64555> https://www.google.com/search?q=-10+modulo+12&ie=utf-8&oe=utf-8
[16:32:23] <z64555> eh, lemme use an example that's more applicable
[16:33:05] <z64555> https://www.google.com/search?q=-10+modulo+12&ie=utf-8&oe=utf-8#q=-10+modulo+360
[16:33:24] <z64555> that's nifty
[16:36:52] <z64555> So, for a PID loop, it would be
[16:37:04] <z64555> error % (range) - (range/2)
[16:37:23] <z64555> that'll shift it into the -180 to 180 range
[16:37:37] <z64555> and then for output/readability, you can +(range/2)
[16:37:46] <z64555> ehh...
[16:37:55] <deshipu> easier to do 0-360
[16:38:04] <deshipu> and then subtract 180 at the end
[16:38:21] <z64555> I guess it depends on your application
[16:38:58] <z64555> for an aerial robot stabelization, it would be -180 to 180, but for a compass it would be 0 to 360
[16:39:44] * z64555 grumbles at horse puns for corrupting the word "stabilization "
[16:40:11] <justanotheruser> what do you call a device that attaches a gear to a structure causing the structure to move
[16:40:34] <branjb> mechanical
[16:40:51] <justanotheruser> where can I buy a mechanical
[16:41:11] * z64555 not sure if serious
[16:41:20] <justanotheruser> ?
[16:41:40] <justanotheruser> Obviously that's not what it's called, but if he could provide a link for what I descsribed, I don't care what he calls it
[16:42:33] <z64555> just drill 3 holes through the gear and attach the gear to the structure using machine screws
[16:43:45] <justanotheruser> is that really the best solution?
[16:44:19] <z64555> Pretty much, it allows as much torque transfer to the structure as you can get
[16:44:26] <justanotheruser> is that what typicall happens?
[16:44:55] <z64555> If you locked the gear into what's normally the axle, you'd get a gear reduction
[16:45:13] <z64555> the diameter of the axle acting as the second gear
[16:45:30] <z64555> or, uh. the axle acting as the second gear
[16:45:33] <z64555> Anyway
[16:45:55] <justanotheruser> hmm
[16:46:26] <z64555> If you drill 3 holes in a triangle that is concentric to the gear, you'll also create a "gear" but with a ratio closer to 1:1
[16:47:35] <z64555> You can make more holes for more screws, if you are concerned about shearing off the screws
[16:47:48] <z64555> but you need at least 3
[16:50:43] <justanotheruser> z64555: I think what I want is a gear attached to an axle going through two mounted ball bearings, then attached to the axle the torque
[16:50:47] <justanotheruser> what do you think?
[16:53:22] <z64555> that'll work too, depends on what the rest of the mechansim will look like
[16:53:56] <justanotheruser> what do you mean?
[16:54:00] <z64555> and how much torque you'd be working with
[16:54:28] <justanotheruser> Is there some mechanism for mounting an axle?
[16:54:42] <justanotheruser> The reason I want the support of two mounted ball bearinsg is I will be dealing with a lot of torque
[16:54:48] <z64555> pillow block bearings, basically
[16:55:21] <z64555> Ok, so this is a high-torque application, I recommend you affix the gear to the armature
[16:55:55] <z64555> have a hole in the armature to allow an axle to pass through both the gear and the armature, which would both spin freely
[16:56:16] <z64555> then mount the axle to the frame using a pair of pillow blocks
[16:56:29] <z64555> or captive bearings
[16:57:09] <z64555> Pillow blocks screw into the side of the frame, the captive bearings go through the frame
[16:57:27] <justanotheruser> what does an armature do?
[16:58:00] <z64555> that's the term for a moving/rotating structure
[16:58:18] <z64555> a stator is a structure which does not move, like with electric motors
[16:58:23] <justanotheruser> so in my design, the axle and load would have been the armature?
[16:58:55] <z64555> yes
[16:59:22] <z64555> unless I'm confusing my terms again
[16:59:26] <z64555> Yeah, I meant to say rotor
[16:59:40] * z64555 is not a mechanical engineer
[17:01:05] <justanotheruser> z64555:
http://imgur.com/a/WmEjF
[17:03:31] <justanotheruser> if that is an accurate representation of your idea, it is similar to mine. so the question becomes how do I mount the axle to the load
[17:04:38] <z64555> weld it, clamp it, or screw it
[17:04:41] <z64555> pick one
[17:05:06] <justanotheruser> are there axles that are designed to be mounted in this way?
[17:05:11] <z64555> no
[17:05:12] <justanotheruser> I feel like I would screw it up
[17:05:33] <justanotheruser> also I'd have to travel a ways to get to a place I can weld at
[17:07:15] <z64555> so make a clamp, or drill through the axle and load structure
[17:07:44] <z64555> the latter I don't recommend, since axles tend to be very hard
[17:07:50] <justanotheruser> hmm
[17:08:02] <justanotheruser> is there some adhesive that should work?
[17:08:12] <justanotheruser> I don't know how much torque a clamp can hold
[17:08:18] <justanotheruser> and drilling takes a few miles as well
[17:08:39] <z64555> a clamp works with friction
[17:09:01] <justanotheruser> right, but higher torque makes it more likely to slip, no?
[17:09:10] <z64555> yes
[17:09:27] <justanotheruser> so do you think theres some adhesive I could use?
[17:09:30] <z64555> So you make a clamp that has more surface contact with the axle
[17:09:39] <z64555> or make more clamps
[17:09:41] <justanotheruser> hmm
[17:09:46] <justanotheruser> true
[17:09:50] <z64555> or
[17:09:59] <z64555> You use the method I suggested earlier :)
[17:10:10] <justanotheruser> drilling or welding?
[17:10:15] <z64555> drilling
[17:10:22] <justanotheruser> right
[17:10:28] <z64555> affixing the gear directly to the load structure
[17:10:56] <z64555> You can also use adhesive in this instance, if the gear face is mostly flat
[17:11:00] <justanotheruser> well I don't know that thats possible
[17:11:15] <justanotheruser> Going straght to load structure without axle
[17:11:58] <z64555> Is the load mostly parallel to the axle or perpendicular?
[17:12:45] <justanotheruser> did you see my drawing?
[17:13:01] <z64555> so it's a paddle, and not a blade?
[17:13:16] <justanotheruser> huh?
[17:13:27] <z64555> paddle, like with paddlewheels
[17:13:28] <justanotheruser> the load is two beams perpendicular to the axle
[17:13:34] <z64555> blades, like with lawn mowers and helicopters
[17:13:52] <justanotheruser> the drawing of the load is just meant to represent its a load
[17:14:24] <z64555> "two beams perpendicular to the axle" ok so go with clamps on both beams
[17:14:37] <justanotheruser> hmm
[17:15:04] <z64555> Do you know the physical properties of the beams and the load?
[17:15:22] <z64555> dimensions and weight, namely
[17:15:50] <justanotheruser> well the max torque is around 80lb-ft
[17:16:31] <justanotheruser> 3-4ft
[17:17:12] <z64555> Ok, so you need a clamp design that can provide at least 40lb-ft of frictional torque
[17:17:27] <z64555> and have a clamp on both beams
[17:17:50] <justanotheruser> drilling sounds more realistic for that weight
[17:18:38] <z64555> drilling will weaken the axle, assuming that you can drill through it
[17:22:00] <justanotheruser> what is the standard way to do this?
[17:22:12] <justanotheruser> what does a bike do?
[17:22:34] <z64555> screw the cog directly to the crank
[17:23:00] <justanotheruser> what does a car do
[17:23:19] <z64555> it uses belts to a gear/axle system
[17:23:32] <justanotheruser> I mean how do the axles connect to the tires
[17:24:04] <z64555> they have a hub that is a sleeve which is welded onto the axle
[17:24:11] <z64555> It slips over the axle
[17:24:22] <justanotheruser> ok so it is friction + welding?
[17:24:29] <z64555> yes
[17:24:41] <justanotheruser> hmm
[17:24:52] <z64555> some of them have a pin that goes through both, as well
[17:24:56] <justanotheruser> I'm surprised there isn't an axle that does this
[17:25:09] <z64555> However, car axles are much larger than the axle that you are working with
[17:25:16] <justanotheruser> yes
[17:27:35] <z64555> so you're stuck with either welding or with a clamp
[17:28:54] <justanotheruser> what about a mounting collar
[17:29:17] <z64555> and how are you going to mount the collar to the axle?
[17:29:41] <justanotheruser> clamping screw
[17:29:51] <z64555> so it's a clamp. only more crappy
[17:30:05] <justanotheruser> what type of clamp do you recommend?
[17:30:52] <z64555> Similar to the collar, but the collar is split and the screw tighens/loosens the collar
[17:31:13] <justanotheruser> what is it called?
[17:31:23] <z64555> I just call it a clamp
[17:31:27] <justanotheruser> oh
[17:31:37] <z64555> You may have to fabricate one yourself
[17:31:50] <justanotheruser> oh my
[17:33:03] <z64555> Looks like some companies call them a "clamping collar"
[17:33:23] <justanotheruser> how do you go from the collar to the load?
[17:33:33] <z64555> weld the collar to the beam
[17:33:53] <z64555> or drill through it and screw it into the beam
[17:34:26] <justanotheruser> why not just weld straight to the beam
[17:34:42] <z64555> b/c you said you couldn't weld :P
[17:35:32] <justanotheruser> but you said weld the collar to the beam
[17:35:51] <z64555> *or drill through it and screw it into the beam
[17:36:16] <justanotheruser> so you're saying use the collar because I can drill throguh that and that won't weaken the axle?
[17:36:43] <z64555> yes
[17:36:54] <z64555> But, if you can weld, then by all means do that
[17:37:03] <z64555> it would be much simpler to fabricate
[17:37:12] <z64555> *weld the beam directly to the axle
[17:39:52] <justanotheruser> hmm
[17:40:21] <justanotheruser> I suppose
[17:40:44] <justanotheruser> I also need to cut an axle shape into my beams
[17:41:07] <justanotheruser> drill a hole into it
[17:44:18] <justanotheruser> z64555: do you think I could drill an axle shape right into the t-slotted framing?
[17:44:51] <z64555> With a drill press, a vise, and a good drill bit, yes
[17:45:05] <justanotheruser> I have that in the same place the welder is
[17:45:11] <z64555> hold up. T-slotted?
[17:45:17] <z64555> Is the beam aluminum?
[17:45:21] <justanotheruser> yes
[17:45:26] <z64555> then you can't weld
[17:45:32] <justanotheruser> oh :O
[17:45:51] <z64555> welding must be done with metals of same or very similar composition
[17:46:04] <z64555> and aluminum welding requires a special machine
[17:46:09] <justanotheruser> hmm
[17:46:22] <z64555> back to clamps
[17:46:26] <justanotheruser> nah
[17:47:30] <justanotheruser> I could mount steel framing to it then mount the steel framing to the aluminum framing
[17:47:50] <z64555> yes, this is possible
[18:22:31] <justanotheruser> hmm
[18:37:59] <justanotheruser> whats the typical mechanism for mounting motor to to an axle?
[20:10:25] <z64555> The axle is either part of the motor's armature/rotor
[20:10:41] <z64555> or is coupled to the rotor shaft
[20:11:13] <z64555> or uses a belt or gear train to transfer power to the axle
[23:41:15] <rue_shop3> if I redo my old cnc machine, the motor I'd have to replace has a 30mm bolt pattern, which dosn't seem to match anything
[23:46:23] <rue_shop3> I'm running 20W into those motors....hmm
[23:49:29] <rue_shop3> the big steppers I have on my newer machine are only 7.3w
[23:52:14] <rue_shop3> what can those cheap little reprap drivers do?
[23:53:54] <rue_shop3> 1A
[23:53:55] <rue_shop3> k
[23:53:58] <rue_shop3> hmm
[23:54:09] <rue_shop3> I have a bunch of 3A drivers from printers
[23:57:47] <rue_shop3> the 8825 can do 1.5A