From jterrell at personhospital.com Mon Dec 4 09:19:57 2006 From: jterrell at personhospital.com (Jeff Terrell) Date: Tue Dec 5 09:22:43 2006 Subject: [WheelCommander] Quadrature Encoders Message-ID: Hey, I'm working on an autonomous robot and I'm wanting to use the Wheel Commander... My setup is: 2) DC Gearhead motors 2) MD03 H-Bridges from Acroname 2) US Digital 256 rotary quadrature encoders (not in motor) in line with belt drive. I talked to Peter about an firmware update for the encoders to allow me to use the US Digital ones... *My question is...I was looking at the Wheel Watcher and noticed that it uses "decoded Clock and Direction signals " ...that being the case...once the encoder firmware is updated will I also need to supply decoded Clock and Direction signals? The US Digital encoders I have just put out raw data... Thanks Jeff Terrell ======================================= CONFIDENTIALITY NOTICE: The information contained in this electronic message is confidential and is intended for the exclusive use of the individual(s) and entity(ies) named as recipients in the message. If you are not an intended recipient of this message, or if you were listed in error, please notify me immediately and delete the material from any computer you may have downloaded it to. You are hereby provided notice that any disclosure, distribution, or copy of this message is strictly prohibited. Thank you. From peters at nubotics.com Tue Dec 5 09:32:02 2006 From: peters at nubotics.com (Pete Skeggs) Date: Tue Dec 5 09:32:11 2006 Subject: [WheelCommander] Quadrature Encoders In-Reply-To: References: Message-ID: <4575AD12.5080406@nubotics.com> Jeff Terrell wrote: > *My question is...I was looking at the Wheel Watcher and noticed that it > uses "decoded Clock and Direction signals " ...that being the case...once the > encoder firmware is updated will I also need to supply decoded Clock and Direction signals? The US Digital encoders I have just put out raw data... > Hi Jeff, If the frequency of the raw data signals (ChA, ChB) from the US Digital encoders is low enough, then the software quadrature decoding mode of the WheelCommander can be used. This is another new feature we are working on, in addition to providing programmable encoder resolution. When we are ready to ship this firmware update, we will provide maximum allowed frequency information. But our rough estimate at this time is that we should be able to handle between 1000 and 10,000 pulses per second, including rising and falling edges of each channel. So, that would equate to between 250 and 2500 encoder lines per second maximum. If the US Digital encoder is on the motor shaft before the gear train (or belt drive?), it may be too fast for our system to function, depending on the resolution of your encoder (in your case, 256) and the RPMs of the motor. Please explain what rotation rate you expect the encoder to run at. Sincerely, Pete Skeggs Nubotics Tech Support From peters at nubotics.com Thu Dec 7 19:59:38 2006 From: peters at nubotics.com (Pete Skeggs) Date: Thu Dec 7 19:59:40 2006 Subject: [WheelCommander] RE: WheelCommander Digest, Vol 2, Issue 1 Message-ID: <4578E32A.60903@nubotics.com> Hi, Jeff, For some reason your messages are being discarded by the email list system, though I get dicarded copy as an administrator of the list. I just made some changes to the content filtering rules to hopefully clear up the problem. Anyway, to answer your question... > I'm not really sure how to figure speed based on mechanics but here is > what I'm working with maybe it would give you an idea...or maybe you > could send me the formula for computing the speed variables... > > The average speed of the robot would be about .5 - 1.5 mph. > > The encoder is in-line between pulleys. > > 1. Wheel base = 15" in diameter > > 2. Wheel size = 4" in diameter > > 3. Wheel timing pulley = 3" in diameter > > 4. Motor timing pulley = .9" in diameter > Well, first, convert mph to inches per second: X (mi/h) * (1h/3600s) * (5280 ft/mi) * (12 in / ft) = X * 17.6 in/sec With a wheel size of 4 inches diameter, the circumference is 4 * pi = 12.6 inches So, the wheels rotate between: 0.5mph * 17.6 * (1 rotation / 12.6 inches) = 0.5 mph * 1.4 = 0.7 rotations / sec to 1.5 mph * 1.4 = 2.1 rotations / sec The relationship between wheel and motor rotations is related to the size of the pulleys. motor rotations / sec * (0.9 / 3) = wheel rotations / sec, or: wheel rotations / sec * (3 / 0.9) = motor rotations / sec So, the motor rotates between: 0.7 * (3 / 0.9) = 0.7 * 3.33 = 2.33 rotations / sec and: 2.1 * 3.33 = 7.0 rotations / sec Now, the question is, how is the encoder driven by the belt "in-line between pulleys"? Is it on its own pulley? If so, what is the diameter of that? If not, how is it driven by the pulleys? With that, we can calculate the estimated rotation rate of the encoder, and so knowing the encoder resolution is 256, we can figure out the signal frequencies that the WheelCommander would need to handle. -Pete From jst20 at hotmail.com Thu Dec 14 15:55:04 2006 From: jst20 at hotmail.com (Jeff Terrell) Date: Thu Dec 14 15:55:16 2006 Subject: [WheelCommander] Encoder pulley placement Message-ID: Skipped content of type text/html From jst20 at hotmail.com Tue Dec 19 13:45:33 2006 From: jst20 at hotmail.com (Jeff Terrell) Date: Tue Dec 19 13:45:47 2006 Subject: [WheelCommander] Encoder pulley placement Message-ID:
Pete,
 
I've sent this a couple of times and not sure if it is getting through...I see where part of it posted to the forum.... sending to a couple of e-mail address hopefully will get though...
 
The encoder pulley is situated between the Wheel and the motor. The belt is a double sided timing belt. The pulley is .7" in diameter. The encoder pulley is in the drive belt loop between the wheel pulley and the motor pulley
 
Thanks for the help
 
Jeff Terrell
 

Get live scores and news about your team: Add the Live.com Football Page From plskeggs at noeticdesign.com Tue Dec 19 15:04:36 2006 From: plskeggs at noeticdesign.com (Pete Skeggs) Date: Tue Dec 19 15:04:30 2006 Subject: [WheelCommander] RE: Encoder pulley placement In-Reply-To: <4578E32A.60903@nubotics.com> References: <4578E32A.60903@nubotics.com> Message-ID: <45887004.7010101@noeticdesign.com> >> The relationship between wheel and motor rotations is related to the >> size of the pulleys. >> motor rotations / sec * (0.9 / 3) = wheel rotations / sec, >> or: >> wheel rotations / sec * (3 / 0.9) = motor rotations / sec >> So, the motor rotates between: >> 0.7 * (3 / 0.9) = 0.7 * 3.33 = 2.33 rotations / sec >> and: >> 2.1 * 3.33 = 7.0 rotations / sec >> Now, the question is, how is the encoder driven by the belt "in-line >> between pulleys"? Is it on its own pulley? If so, what is the >> diameter of that? If not, how is it driven by the pulleys? > The encoder pulley is situated between the Wheel and the motor. The > belt is a double sided timing belt. The pulley is .7" in diameter. The > encoder pulley is in the drive belt loop between the wheel pulley and > the motor pulley > OK, so, motor rotations / sec * (0.9 / 0.7) = encoder rotations / sec encoder rotations / sec * (0.7 / 0.9) = motor rotations / sec so the encoder rotates between: 2.33 * (0.9 / 0.7) = 3 rotations / sec and: 7.0 * (0.9 / 0.7) = 9 rotations / sec >> With that, we can calculate the estimated rotation rate of the >> encoder, and so knowing the encoder resolution is 256, we can figure >> out the signal frequencies that the WheelCommander would need to handle. At 256 encoder pulses per rotation, this is then between: 3 * 256 = 768 Hz. and 9 * 256 = 2304 Hz. Our firmware processes an interrupt for each edge of each of the two encoder channels, so this ends up being an interrupt frequency 4 times higher, or between 3072 Hz and 9216 Hz. This may work, but will require testing and evaluation here. Sorry for the long delay in replying. Sincerely, Pete Skeggs Nubotics Tech Support From jst20 at hotmail.com Sat Dec 23 18:09:12 2006 From: jst20 at hotmail.com (Jeff Terrell) Date: Sat Dec 23 18:09:20 2006 Subject: [WheelCommander] Encoders continued Message-ID:
Pete,
 
You say this may work...do you mean that it will work as is...or after the firmware update?
 
Thanks
 
Jeff

Get FREE Web site and company branded e-mail from Microsoft Office Live From peters at nubotics.com Sun Dec 24 21:27:12 2006 From: peters at nubotics.com (Pete Skeggs) Date: Sun Dec 24 21:27:16 2006 Subject: [WheelCommander] Encoders continued In-Reply-To: References: Message-ID: <458F6130.6040302@nubotics.com> Jeff Terrell wrote: > You say this may work...do you mean that it will work as is...or after > the firmware update? Hi Jeff, Not as is, after the firmware update. The circuit was not originally designed for this application, and so it is stretching the processor resources to do this. I hope to get some time to work on this soon. My current client is taking all of my time, as they are trying to get a product ready to demo at the Consumer Electronics Show in Las Vegas in about 2 weeks. Happy Holidays! -Pete