Hey - for you early adopters, could you please follow the assembly instructions and provide feedback here for any suggestions that you might have?
We want to make the documentation strike a happy balance between *complete newbie* and "old pro* (if there is such a thing).
@ChrisL8 - Hope you document your build and give us a report again for the power distribution board! If not, I understand - it is a lot of work to do what you did with the Arlo.
@ChrisL8 - Hope you document your build and give us a report again for the power distribution board! If not, I understand - it is a lot of work to do what you did with the Arlo.
I started on it Friday night, finished it Saturday morning and was using it Saturday evening and Sunday! It went fast and well.
I did take some pictures and notes so I'll throw up a short write up when I have some free time.
... someone who is by no means a soldering expert.
Ditto to that. Was a bit disappointed to see that this was a kit and not an assembled product. I still might pick one up during the next free shipping or sale event though...
Ditto to that. Was a bit disappointed to see that this was a kit and not an assembled product. I still might pick one up during the next free shipping or sale event though...
It's a fast assembly. I just built one with a 12-year-old in an hour.
Here's the situation. If we program our selective solder to build these then the costs only go up, the release is delayed, etc. With a low volume product it's just not cost effective for us to try to amortize manufacturing setups associated with soldering machines. We'd probably have to charge around $150 and I don't think you'd want to pay that much. At least that's the guess I made.
Seems like half the fun would be soldering the kit together. A fun way to become intimate with the Schematics, And you still end up with a quality item.:thumb:
Seems like half the fun would be soldering the kit together. A fun way to become intimate with the Schematics, And you still end up with a quality item.:thumb:
I think so too.
There is an alternative for people who don't want to solder. Just ask one of our tech support staff to put it together and pay them fifty bucks to build it on a Saturday. The frequency of such requests are really low so I wouldn't see it as a conflict with their real profession.
That wasn't a criticism - I totally get economics of the situation. I have no doubt she is a quality parallax product. My soldering skills are workable at this point, but I just take no joy in it. And yea, I will buy one eventually anyway.
Especially when you can still buy fully-leaded solder. :thumb:
I'm glad to hear you say that! As a soldering novice I thought maybe it was just my lack of skill that made me heavily favor using leaded solder. I find the lead free kind very difficult to use. In fact I used up my last bit of leaded solder on the power distribution board.
I thought I'd share something I discovered while using the Arlo Power Distribution Board (APDB) with the Activity Board (AB).
If the AB is connected to the APDB through one of the "6.5V Aux. Power" screw terminals while the AB is also connected to a PC with its USB port then the power from the USB port get routed into the APDB. The APDB draws a lot of current from the USB connection. The combined boards pulled more than 450mA from the USB port. The vast majority of this power was drawn by the APDB.
My present solution is to disconnect the positive wire leading from the APDB to the AB. I'm leaving the ground wire connected.
I had the main switch of the APDB turned off while the AB was connected to the USB. I don't know how the current flow over the USB would change if the main power switch of the APDB were turn on (it didn't seem like it would be a good idea to try it).
I think the high current draw keep the AB in reset when I attempted to connect my recently built robot to a notebook computer. Besides being concerned the high current draw causes problems for the Propeller board, I'm also concerned the reverse current through the switch regulator could be a problem.
I'll probably add a switch between the APDB and the AB but it would be nice if the APDB took care of this for us.
Another concern I have about the APDB is the fuse rating. 5A to each motor doesn't seem like a lot and a "TURN" command issued through the Eddie software, promptly blew both 5A fuses (the robot was on blocks and the wheels were free to spin). While preventing abrupt speed changes is something I plan to update in the Eddie software, I also think it's not unreasonable to expect these motors to pull over 5A and I think the fuses should be adjusted upward from the 5A ratings.
I think 10A fuses should be used where there are now 5A fuses and I also think the 10A fuses should be changed to 20A fuses.
Neither of these issues were really a big deal to me but I think these issues could be very frustrating to someone new to robotics.
BTW, This board is beautiful! The switches have really nice rubber seals around them which look fantastic. AND the switches light up!
Doesn't the Arlo Power Distribution Board look nice?!
I hate to think what my bot would look like without the Arlo Power Distribution Board. I obviously need all the help I can get managing wires.
I do plan to clean up the looks of he bot once I get a few of the OPP #8 software goals accomplished. I used a piece of expanded PVC I had on hand to make a bot with the same wheel spacing as the Arlo.
Edit: As I think about this a bit, I think one could probably just leave the AB powered from the on board PC without any connection directly to the APDB. All the devices the AB connects with will have a common ground so the AB doesn't need a ground connection with the APDB.
Thanks Duane - I figured you'd quickly mount these up on something and roll your own bot.
I agree on the motor fuses - 10A is more like it. Our new Dual H-Bridge 10A product is indeed "10A" so this makes sense. More on that board later. I'll have MattG take a look at your additional points to see if there's anything else we might want to consider.
I'll be sporadically around this week but will check in on the Eddie => Arlo thread when I can.
I thought I'd share something I discovered while using the Arlo Power Distribution Board (APDB) with the Activity Board (AB).
If the AB is connected to the APDB through one of the "6.5V Aux. Power" screw terminals while the AB is also connected to a PC with its USB port then the power from the USB port get routed into the APDB. The APDB draws a lot of current from the USB connection. The combined boards pulled more than 450mA from the USB port. The vast majority of this power was drawn by the APDB.
My present solution is to disconnect the positive wire leading from the APDB to the AB. I'm leaving the ground wire connected.
I had the main switch of the APDB turned off while the AB was connected to the USB. I don't know how the current flow over the USB would change if the main power switch of the APDB were turn on (it didn't seem like it would be a good idea to try it).
I think the high current draw keep the AB in reset when I attempted to connect my recently built robot to a notebook computer. Besides being concerned the high current draw causes problems for the Propeller board, I'm also concerned the reverse current through the switch regulator could be a problem.
I'll probably add a switch between the APDB and the AB but it would be nice if the APDB took care of this for us.
Another concern I have about the APDB is the fuse rating. 5A to each motor doesn't seem like a lot and a "TURN" command issued through the Eddie software, promptly blew both 5A fuses (the robot was on blocks and the wheels were free to spin). While preventing abrupt speed changes is something I plan to update in the Eddie software, I also think it's not unreasonable to expect these motors to pull over 5A and I think the fuses should be adjusted upward from the 5A ratings.
I think 10A fuses should be used where there are now 5A fuses and I also think the 10A fuses should be changed to 20A fuses.
Neither of these issues were really a big deal to me but I think these issues could be very frustrating to someone new to robotics.
BTW, This board is beautiful! The switches have really nice rubber seals around them which look fantastic. AND the switches light up!
Doesn't the Arlo Power Distribution Board look nice?!
I hate to think what my bot would look like without the Arlo Power Distribution Board. I obviously need all the help I can get managing wires.
I do plan to clean up the looks of he bot once I get a few of the OPP #8 software goals accomplished. I used a piece of expanded PVC I had on hand to make a bot with the same wheel spacing as the Arlo.
Edit: As I think about this a bit, I think one could probably just leave the AB powered from the on board PC without any connection directly to the APDB. All the devices the AB connects with will have a common ground so the AB doesn't need a ground connection with the APDB.
Duane,
I love this design (like the Singay Robot)
Any more photos?
Is the code Eddie=>Arlo reay? I would like to test it on my Robot.
As I think about this a bit, I think one could probably just leave the AB powered from the on board PC without any connection directly to the APDB
True, but doesn't that only solve the issue IF there's a PC onboard (which in the case of an "Eddie" application would be ok) - and then we are we saying that anytime we connect to the AB USB, we're back-feeding 450 milliamps into Arlo from any PC, right?
It's almost like we should add an output diode from the regulator and then alter the regulator output to accommodate the vdrop to eliminate the back-feed condition in all cases then?
then we are we saying that anytime we connect to the AB USB, we're back-feeding 450 milliamps into Arlo from any PC, right?
I'm not sure of the exact amount of current the APDB pulled from the AB. The 450mA figure includes the power used by the AB. The amount pulled by the APDB is probably closer to 300mA since the AB draws about about 150mA on its own. It's possible the APDB could pull more than 300mA but the current was limited by the USB hub.
Those USB power meters erco made me purchase sure have been handy.
It's almost like we should add an output diode from the regulator and then alter the regulator output to accommodate the vdrop to eliminate the back-feed condition in all cases then?
This seems like a good plan. I agree this is something worth fixing. I think it's reasonable to assume people will want to both power their AB from the APDB as well as occasionally connects a USB cable to the AB while the AB is attached to the APDB.
I didn't test to see what would happen if the APDB were turned on with the USB connected. What happens if the motor switch is turned on while the main switch is off? I have a horrific scenario in mind of an Arlo attempting to power its motors from the AB's USB connection.
Did I mention the board looks nice? Now I need to spruce up the rest of my robot so the nice wheels and APDB don't look out of place.
@ Matt, Duane, and Ken - Alright folks - Got an APDB on the way (my Christmas prize!), Also ordered a pair of HB-25 Mounting Bracket(s) and Assorted 3-Wire Extension Cable 30-Pack Thanks for the 10% off, and the free shipping and the free WS2812BRGB LED Module Parallax! What a deal!
Been wanting one of these for a while... Now - if I can just wait till Christmas...
I thought I'd give this thread a bump to see if the APDB uses different fuses yet.
The APDB is a nice product but it costs a lot. IMO, customers should expect it to work well without having to purchase (many) extra components.
I think this thread shows I'm not the only one who had problems with the 5A fuses.
First I've heard of the issue from my customers though I admit I've experienced the problem of the 5A fuses blowing too easily.
What you're suggesting is that we provide 10A fuses, right? We can do that - I can't think of a reason we wouldn't do that.
The kit has 2-10A, 2-5A, 3-2A and 1-1A fuses. I'll need to take a look to see if the 2-5A fuses are for the motors, or if we've already addressed this improvement.
I guess I don't have any angst in changing those to 10 amp fuses. I don't think that we've ever blown any here (I know I haven't) - even during our load tests.
-Matt
PS If you want some 10 ampers - PM me with your mailing address.:D
-MattG .
I don't think that we've ever blown any here (I know I haven't) - even during our load tests.
You were probably using code which treated the motors kindly.
I should clarify, when I blew the 5A on my board, it was when I was first testing the Eddie firmware with the HB-25 controllers. I had the robot up on blocks with the wheels free to spin but my control algorithm was so out of whack the motor oscillated back and forth for a fraction of a second before both 5A fuses blew.
It didn't take me long to find some replacement fuses but I thought it was strange to have a 5A fuse going to a 25A motor controller and motors rated up to 8A.
If it's unusual for the 5A fuses to blow, then maybe they are appropriate and I overreacted when the fuses blew on my robot. I think having the fuses blow in less than a second, when first trying the motors, added to my frustration. In hindsight, I should have tried some nice PropBOE-bot code which I already knew to work.
Another contributing factor was one of my motors had something misaligned inside the gearbox. It drew over 2A on the bench with the wheel free to spin. After pulling the motor and worm drive from the gearbox and realigning the worm drive mechanism, the motor's current consumption was reduced.
Initially seeing 2A of current with the motor free to turn made me more inclined to think the 5A fuses were not appropriate.
I hope you take this extra information into account before deciding to switch fuses.
I still thing the current pull from the PAB's USB line while the distribution board is turned off should probably be addressed.
I'm not up on the full details of everything. I was just pointed to this thread by Matt G. What I did want to say is that given the nominal current draw of <2 amps typical and a stall current of 8-10 amps, then a 5 amp fuse is the closest standard fuse and is appropriate. If you can get a 7 or 8 amp fuse then by all means, use that, however if you put a 10 amp fuse in it defeats the purpose of what the fuse is there to do, which is protect the motors and other circuitry from being damaged during stall current conditions. I guess the important thing to do here is properly characterize the normal/typical operating conditions before making any changes.
I just tested every motor that we currently have in stock...Two were defective - they drew over double what they should have under no-load conditions. (They've been pulled from our sales inventory) Moreover, they rapidly increased in current consumption when subjected to even a minor load scenario.
Every other motor tested fine (somewhere between point 7 and 1.2 amps no-load), With moderate load applied they went up to about 2 amps. Therefore I believe at this point that the 5amp fuse IS appropriate - provided the motors are operating to their specs.
Guys, confirm for me - which motor (L or R) was blowing your fuse?
Please PM me with your shipping addresses and I'll get some known-good motors out to you pronto
Such a motor could easily be pulling 4A continuous in normal use with a light load without stalling.
erco,
I understand that. But I don't believe it should be. I think that is one of the motors like Matt G is discussing (I saw his reply while I was typing this). It still doesn't change the fact that if you put too high of a fuse in there you won't have any protection. Once your fuse rating exceeds the maximum current demand it is almost useless. I say almost because it might protect against a direct short if the P/S can supply that kind of current.
P.S. - This is one of those newly discovered things we're trying to address. I just want to make sure we address it in the right manner.
Guys, confirm for me - which motor (L or R) was blowing your fuse?
Both fuses blew within a second of my first turning on the robot. As I mentioned, the code I was using hadn't been corrected to account for the HB-25 power levels and the strong oscillation caused the fuses to blow.
I suppose the fuses were doing there job and should have blown with the oscillating motors.
The problem motor was on the right side. At the time, the high current required to turn the right motor without a load contributed to my thoughts of improperly rated fuses.
As I mentioned above, I think I over reacted in thinking the fuses were not appropriate.
Please PM me with your shipping addresses and I'll get some known-good motors out to you pronto
I'll take you up on the motor. I'll probably wait to replace the right motor until I have the Eddie firmware working the way I want.
The right motor requires about 1.5A current without a load but even with this slightly out of spec motor, my robot behaves pretty well.
This is the same robot I used to drive my rectangular encoder test course in my livingroom as shown in the video attached to post #41 of the OPP#8 thread. IMO, it works well.
I drove the same course after adding the diagnostic LEDs shown in post #48 but the LEDs showed very little difference in the position error between the left and right wheels. In order to produce enough position error to measure, I had to resort to spinning the robot at top speed (as shown in post #48). At high speeds there's a small position error with the right motor consistently slower than the left motor.
While I'll want to eventually replace the right motor, I think having a slightly out of spec motor could be a benefit when testing the motor control code. IMO, the control algorithm should compensate for the mismatched motor.
I still have plans to refine the Eddie firmware a bit more. I think I can improve the position control as the robot stops and I would also like to improve the control at high speeds. Using the slower motor should be a good test for an improved algorithm.
All MMWK's currently in stock have been checked and are good to go.
All Left and Right Motors (raw materials) that we have in stock have been tested. The big identifier of a defective unit ended up being ones that pulled in excess of 2 amps under no load at 12vdc. "Good" motors pull about 1 to 1.4 amps under no-load. Under "driving-around" conditions, the amperage goes up only slightly. But on motors that pulled 2 amps under no-load, the amperage shoots up quickly (Naturally, Duane would go and get one of those .. Could be lack of grease or poor worm assembly...doesn't matter they're landfill - er...I mean recyclables now. .
Regardless, all good motors and full Motor Mount and Wheel Kits have been returned to stock and are anxious to be adopted into loving homes where they will be cherished for many years to come :-)
Oh, and we've instigated an incoming inspection/test on these, before they're allowed to go into stock.
-MattG
PS PM me with your address if you would like a new motor replacement.
Comments
Hey - for you early adopters, could you please follow the assembly instructions and provide feedback here for any suggestions that you might have?
We want to make the documentation strike a happy balance between *complete newbie* and "old pro* (if there is such a thing).
-MattG
It has arrived! I should have some feedback by Monday from someone who is by no means a soldering expert.
I started on it Friday night, finished it Saturday morning and was using it Saturday evening and Sunday! It went fast and well.
I did take some pictures and notes so I'll throw up a short write up when I have some free time.
Ditto to that. Was a bit disappointed to see that this was a kit and not an assembled product. I still might pick one up during the next free shipping or sale event though...
It's a fast assembly. I just built one with a 12-year-old in an hour.
Here's the situation. If we program our selective solder to build these then the costs only go up, the release is delayed, etc. With a low volume product it's just not cost effective for us to try to amortize manufacturing setups associated with soldering machines. We'd probably have to charge around $150 and I don't think you'd want to pay that much. At least that's the guess I made.
Ken Gracey
I think so too.
There is an alternative for people who don't want to solder. Just ask one of our tech support staff to put it together and pay them fifty bucks to build it on a Saturday. The frequency of such requests are really low so I wouldn't see it as a conflict with their real profession.
Ken Gracey
I'll do it for $49.99 !
But wait, erco! You're the low-price leader. Based on the Ping/HC04 ultrasonic pricing ratio shouldn't you be doing this job for two dollars?
-Matt-Keep-The-Lead-In-G.
I'm glad to hear you say that! As a soldering novice I thought maybe it was just my lack of skill that made me heavily favor using leaded solder. I find the lead free kind very difficult to use. In fact I used up my last bit of leaded solder on the power distribution board.
I thought I'd share something I discovered while using the Arlo Power Distribution Board (APDB) with the Activity Board (AB).
If the AB is connected to the APDB through one of the "6.5V Aux. Power" screw terminals while the AB is also connected to a PC with its USB port then the power from the USB port get routed into the APDB. The APDB draws a lot of current from the USB connection. The combined boards pulled more than 450mA from the USB port. The vast majority of this power was drawn by the APDB.
My present solution is to disconnect the positive wire leading from the APDB to the AB. I'm leaving the ground wire connected.
I had the main switch of the APDB turned off while the AB was connected to the USB. I don't know how the current flow over the USB would change if the main power switch of the APDB were turn on (it didn't seem like it would be a good idea to try it).
I think the high current draw keep the AB in reset when I attempted to connect my recently built robot to a notebook computer. Besides being concerned the high current draw causes problems for the Propeller board, I'm also concerned the reverse current through the switch regulator could be a problem.
I'll probably add a switch between the APDB and the AB but it would be nice if the APDB took care of this for us.
Another concern I have about the APDB is the fuse rating. 5A to each motor doesn't seem like a lot and a "TURN" command issued through the Eddie software, promptly blew both 5A fuses (the robot was on blocks and the wheels were free to spin). While preventing abrupt speed changes is something I plan to update in the Eddie software, I also think it's not unreasonable to expect these motors to pull over 5A and I think the fuses should be adjusted upward from the 5A ratings.
I think 10A fuses should be used where there are now 5A fuses and I also think the 10A fuses should be changed to 20A fuses.
Neither of these issues were really a big deal to me but I think these issues could be very frustrating to someone new to robotics.
BTW, This board is beautiful! The switches have really nice rubber seals around them which look fantastic. AND the switches light up!
Doesn't the Arlo Power Distribution Board look nice?!
I hate to think what my bot would look like without the Arlo Power Distribution Board. I obviously need all the help I can get managing wires.
I do plan to clean up the looks of he bot once I get a few of the OPP #8 software goals accomplished. I used a piece of expanded PVC I had on hand to make a bot with the same wheel spacing as the Arlo.
Edit: As I think about this a bit, I think one could probably just leave the AB powered from the on board PC without any connection directly to the APDB. All the devices the AB connects with will have a common ground so the AB doesn't need a ground connection with the APDB.
I agree on the motor fuses - 10A is more like it. Our new Dual H-Bridge 10A product is indeed "10A" so this makes sense. More on that board later. I'll have MattG take a look at your additional points to see if there's anything else we might want to consider.
I'll be sporadically around this week but will check in on the Eddie => Arlo thread when I can.
Ken Gracey
I love this design (like the Singay Robot)
Any more photos?
Is the code Eddie=>Arlo reay? I would like to test it on my Robot.
It's almost like we should add an output diode from the regulator and then alter the regulator output to accommodate the vdrop to eliminate the back-feed condition in all cases then?
-MattG
I'm not sure of the exact amount of current the APDB pulled from the AB. The 450mA figure includes the power used by the AB. The amount pulled by the APDB is probably closer to 300mA since the AB draws about about 150mA on its own. It's possible the APDB could pull more than 300mA but the current was limited by the USB hub.
Those USB power meters erco made me purchase sure have been handy.
This seems like a good plan. I agree this is something worth fixing. I think it's reasonable to assume people will want to both power their AB from the APDB as well as occasionally connects a USB cable to the AB while the AB is attached to the APDB.
I didn't test to see what would happen if the APDB were turned on with the USB connected. What happens if the motor switch is turned on while the main switch is off? I have a horrific scenario in mind of an Arlo attempting to power its motors from the AB's USB connection.
Did I mention the board looks nice? Now I need to spruce up the rest of my robot so the nice wheels and APDB don't look out of place.
Been wanting one of these for a while... Now - if I can just wait till Christmas...
The APDB is a nice product but it costs a lot. IMO, customers should expect it to work well without having to purchase (many) extra components.
I think this thread shows I'm not the only one who had problems with the 5A fuses.
First I've heard of the issue from my customers though I admit I've experienced the problem of the 5A fuses blowing too easily.
What you're suggesting is that we provide 10A fuses, right? We can do that - I can't think of a reason we wouldn't do that.
The kit has 2-10A, 2-5A, 3-2A and 1-1A fuses. I'll need to take a look to see if the 2-5A fuses are for the motors, or if we've already addressed this improvement.
Will check with Matt and report back.
Ken Gracey
-Matt
PS If you want some 10 ampers - PM me with your mailing address.:D
-MattG .
You were probably using code which treated the motors kindly.
I should clarify, when I blew the 5A on my board, it was when I was first testing the Eddie firmware with the HB-25 controllers. I had the robot up on blocks with the wheels free to spin but my control algorithm was so out of whack the motor oscillated back and forth for a fraction of a second before both 5A fuses blew.
It didn't take me long to find some replacement fuses but I thought it was strange to have a 5A fuse going to a 25A motor controller and motors rated up to 8A.
If it's unusual for the 5A fuses to blow, then maybe they are appropriate and I overreacted when the fuses blew on my robot. I think having the fuses blow in less than a second, when first trying the motors, added to my frustration. In hindsight, I should have tried some nice PropBOE-bot code which I already knew to work.
Another contributing factor was one of my motors had something misaligned inside the gearbox. It drew over 2A on the bench with the wheel free to spin. After pulling the motor and worm drive from the gearbox and realigning the worm drive mechanism, the motor's current consumption was reduced.
Initially seeing 2A of current with the motor free to turn made me more inclined to think the 5A fuses were not appropriate.
I hope you take this extra information into account before deciding to switch fuses.
I still thing the current pull from the PAB's USB line while the distribution board is turned off should probably be addressed.
I'm not up on the full details of everything. I was just pointed to this thread by Matt G. What I did want to say is that given the nominal current draw of <2 amps typical and a stall current of 8-10 amps, then a 5 amp fuse is the closest standard fuse and is appropriate. If you can get a 7 or 8 amp fuse then by all means, use that, however if you put a 10 amp fuse in it defeats the purpose of what the fuse is there to do, which is protect the motors and other circuitry from being damaged during stall current conditions. I guess the important thing to do here is properly characterize the normal/typical operating conditions before making any changes.
Duane's stock motor pulled over 2 amps NO LOAD. http://forums.parallax.com/showthread.php/158701-Motor-Mount-amp-Wheel-Kit-Maintenance-Which-Grease
Such a motor could easily be pulling 4A continuous in normal use with a light load without stalling.
Yep. All over it....
So here's some data:
I just tested every motor that we currently have in stock...Two were defective - they drew over double what they should have under no-load conditions. (They've been pulled from our sales inventory) Moreover, they rapidly increased in current consumption when subjected to even a minor load scenario.
Every other motor tested fine (somewhere between point 7 and 1.2 amps no-load), With moderate load applied they went up to about 2 amps. Therefore I believe at this point that the 5amp fuse IS appropriate - provided the motors are operating to their specs.
Guys, confirm for me - which motor (L or R) was blowing your fuse?
Please PM me with your shipping addresses and I'll get some known-good motors out to you pronto
-MattG
erco,
I understand that. But I don't believe it should be. I think that is one of the motors like Matt G is discussing (I saw his reply while I was typing this). It still doesn't change the fact that if you put too high of a fuse in there you won't have any protection. Once your fuse rating exceeds the maximum current demand it is almost useless. I say almost because it might protect against a direct short if the P/S can supply that kind of current.
P.S. - This is one of those newly discovered things we're trying to address. I just want to make sure we address it in the right manner.
Both fuses blew within a second of my first turning on the robot. As I mentioned, the code I was using hadn't been corrected to account for the HB-25 power levels and the strong oscillation caused the fuses to blow.
I suppose the fuses were doing there job and should have blown with the oscillating motors.
The problem motor was on the right side. At the time, the high current required to turn the right motor without a load contributed to my thoughts of improperly rated fuses.
As I mentioned above, I think I over reacted in thinking the fuses were not appropriate.
I'll take you up on the motor. I'll probably wait to replace the right motor until I have the Eddie firmware working the way I want.
The right motor requires about 1.5A current without a load but even with this slightly out of spec motor, my robot behaves pretty well.
This is the same robot I used to drive my rectangular encoder test course in my livingroom as shown in the video attached to post #41 of the OPP#8 thread. IMO, it works well.
I drove the same course after adding the diagnostic LEDs shown in post #48 but the LEDs showed very little difference in the position error between the left and right wheels. In order to produce enough position error to measure, I had to resort to spinning the robot at top speed (as shown in post #48). At high speeds there's a small position error with the right motor consistently slower than the left motor.
While I'll want to eventually replace the right motor, I think having a slightly out of spec motor could be a benefit when testing the motor control code. IMO, the control algorithm should compensate for the mismatched motor.
I still have plans to refine the Eddie firmware a bit more. I think I can improve the position control as the robot stops and I would also like to improve the control at high speeds. Using the slower motor should be a good test for an improved algorithm.
All MMWK's currently in stock have been checked and are good to go.
All Left and Right Motors (raw materials) that we have in stock have been tested. The big identifier of a defective unit ended up being ones that pulled in excess of 2 amps under no load at 12vdc. "Good" motors pull about 1 to 1.4 amps under no-load. Under "driving-around" conditions, the amperage goes up only slightly. But on motors that pulled 2 amps under no-load, the amperage shoots up quickly (Naturally, Duane would go and get one of those
Regardless, all good motors and full Motor Mount and Wheel Kits have been returned to stock and are anxious to be adopted into loving homes where they will be cherished for many years to come :-)
Oh, and we've instigated an incoming inspection/test on these, before they're allowed to go into stock.
-MattG
PS PM me with your address if you would like a new motor replacement.
These aren't just robot wheels and motors they're little works of art. It's hard to articulate the beauty of these things.
They look every bit as nice in real life as they do in the photos.
I suppose the plastic version look nice but if any of you are thinking of getting a set of these go for the aluminum version. They're really nice.